From paul.mccann at adelaide.edu.au Mon Sep 1 10:50:24 2014 From: paul.mccann at adelaide.edu.au (Paul McCann) Date: Mon, 1 Sep 2014 20:20:24 +0930 Subject: [TxMt] Re: Latex Typeset and View new failure In-Reply-To: <8E4440BE-FE46-42EC-989A-EA35AADDFCBA@textmate.org> References: <72C1A350-893B-4573-BA32-167D5CCFB3B6@textmate.org> <20140831163250.GA14379@Mini-bihan.local> <8E4440BE-FE46-42EC-989A-EA35AADDFCBA@textmate.org> Message-ID: Hi all, looks like Skim is pretty much out-of-action with Yosemite for now: seems pdfkit is misbehaving rather chronically. I saw a couple of instances where opening a simple 200KB pdf caused Skim to first expand to 8GB of memory in Activity Monitor (causing "out of memory: quit applications you're no longer using" warnings), and then needed to be forced to quit. In short: probably not worth chasing too hard until some things in pdfkit get some serious attention. All part of the fun of beta testing 10.10. Cheers, Paul On Mon, Sep 1, 2014 at 2:35 AM, Allan Odgaard wrote: > On 31 Aug 2014, at 18:32, Gildas Hamel wrote: > >> I experienced the same thing after upgrading to Yosemite beta, because >> Yosemite points to Ruby 2.0 >> (/System/Library/Frameworks/Ruby.framework/Versions). I added Ruby 1.8 to >> make it work again. > > > TextMate will automatically fetch ruby 1.8 on Yosemite (if running a command > that requires it). > > So things are supposed to work out-of-the-box on 10.10 without the need to > manually edit or install anything. > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate From 100621.1 at masonsmusic.co.uk Tue Sep 2 11:38:06 2014 From: 100621.1 at masonsmusic.co.uk (Justin Catterall) Date: Tue, 2 Sep 2014 12:38:06 +0100 Subject: [TxMt] Re: Where do I update the definition of a word? In-Reply-To: References: Message-ID: <06505691-DDEC-4578-9A11-626C71416BA0@masonsmusic.co.uk> On 22 Aug 2014, at 08:21, Allan Odgaard wrote: > On 20 Aug 2014, at 22:32, Pedro Melo wrote: > >> using the Perl gramar, if I have the cursor on a package name, like >> Time::Mo|ment, (| is the cursor) and I try to get help, it should try to >> get help on Time::Moment but right now it only uses Moment. >> >> I had this problem with TM1 also but I fixed it by including : in the >> definition of the "word". I don't know if this is still the way to do it on >> TM2 or if there is a better way. > > There is definitely a better way, since changing word characters would affect word movement, buffer completion, and other things that rely on word definitions. > > Looking at the command this is what it does: > > word=${TM_SELECTED_TEXT:-$TM_CURRENT_WORD} > > A quick fix is to change that line to the following: > > word=$(ruby18 -r"$TM_SUPPORT_PATH/lib/current_word" -e"puts Word.current_word(':A-Za-z')") > > This calls the `current_word` function from the bundle support bundle. This function looks at the `TM_CURRENT_LINE` and `TM_LINE_INDEX` variables and then extracts the word surrounding the caret based on the allowed characters (passed as its argument). > > Can you tell me if the above character range would be correct for perl? (':_A-Za-z0-9') Module names may contain digits. > Also, how do I install perl documentation? For me, perldoc reports ?no documentation? for my tests (though the command suppresses that, so it?s effectively a no-op, another thing that I?d like to get fixed). Documentation should be installed automatically with each module. I would do some testing but I don't know where "word=${TM_SELECTED_TEXT:-$TM_CURRENT_WORD}" is to test your quick fix. -- Justin C, by the sea. From realoreocookie at gmx.de Tue Sep 2 16:22:57 2014 From: realoreocookie at gmx.de (Max Lein) Date: Tue, 2 Sep 2014 12:22:57 -0400 Subject: [TxMt] search & replace window borked in alpha.9561 Message-ID: The search and replace window has a serious bug: the text in the search and in the replace text fields is not updated as you type, meaning whatever was in these fields from the last search is displayed properly. But if you edit the text in one of the fields, it is not updated. Functionality-wise, everything still seems to work, meaning that the text you?ve entered blindly in the search field still is what TextMate searches for and the text in the replace text field is what the search term will be replaced with. Also, if you do a search all, the first time you press the button Search All, the Replace All button remains grayed out. Only after pressing Search All a second time does the Replace All button become active. Does anyone else have the same problem? Max From mailinglist at textmate.org Tue Sep 2 19:27:26 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 02 Sep 2014 21:27:26 +0200 Subject: [TxMt] Re: File changed status wrong when modified by a bundle command In-Reply-To: <373CE520-7C57-4E46-A295-ED9C650F5EB5@efrancos.com> References: <20140809165408.534.qmail@f4mail-235-237.rediffmail.com> <08CDBA6E-DE1D-4F24-99E1-FCC7289FF120@textmate.org> <373CE520-7C57-4E46-A295-ED9C650F5EB5@efrancos.com> Message-ID: On 17 Aug 2014, at 10:46, Eduardo Francos wrote: > As I said before the problem occurs only when the version file is > modified by a command AND it's the currently active tab [?] Thanks for the test case, I have fixed the issue in 2.0-alpha.9563 (current nightly build). From mailinglist at textmate.org Tue Sep 2 19:30:50 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 02 Sep 2014 21:30:50 +0200 Subject: [TxMt] Re: search & replace window borked in alpha.9561 In-Reply-To: References: Message-ID: On 2 Sep 2014, at 18:22, Max Lein wrote: > The search and replace window has a serious bug: the text in the > search and in the replace text fields is not updated as you type [?] Does a relaunch fix it? I have heard about the find dialog going into this state, but it?s rare and fixed by a relaunch. Unfortunately I do not know what?s causing it, as no information (like an uncaugt exception) is logged. If anyone have steps to reproduce then I would of course love ot learn about these. > Also, if you do a search all, the first time you press the button > Search All, the Replace All button remains grayed out. Only after > pressing Search All a second time does the Replace All button become > active. I can?t reproduce this with 2.0-alpha.9563 but there have been a lot of changes done to the find dialog, so I assume it has been fixed, otherwise let me know. From mailinglist at textmate.org Tue Sep 2 20:23:49 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 02 Sep 2014 22:23:49 +0200 Subject: [TxMt] Re: Where do I update the definition of a word? In-Reply-To: <06505691-DDEC-4578-9A11-626C71416BA0@masonsmusic.co.uk> References: <06505691-DDEC-4578-9A11-626C71416BA0@masonsmusic.co.uk> Message-ID: <967F3C5A-75B8-439A-AA69-D90CB82281DF@textmate.org> On 2 Sep 2014, at 13:38, Justin Catterall wrote: >> Also, how do I install perl documentation? For me, perldoc reports >> ?no documentation? for my tests (though the command suppresses >> that, so it?s effectively a no-op, another thing that I?d like to >> get fixed). > > Documentation should be installed automatically with each module. > > I would do some testing but I don't know where > "word=${TM_SELECTED_TEXT:-$TM_CURRENT_WORD}" is to test your quick > fix. I pushed an updated documentation? command with these improvements: - If caret is on a module e.g. Pod::Html, then the entire module (with namespace) will be looked up. - If no word is under caret, an input dialog is shown. - If no documentation is found, a tool tip is shown. Your perl bundle should automatically update within the next few hours. For future reference, you can use Bundles ? Select Bundle Item? (??T) and do key equivalent search (?2) for ?H and you?ll find the command responsible for documentation lookup. From this dialog you can click ?edit? (or option return) to view (and edit) the source of the command. ? https://github.com/textmate/perl.tmbundle/commit/391a076fdbcdce2271da57ab3c6e503e2df90407 From realoreocookie at gmx.de Wed Sep 3 12:20:19 2014 From: realoreocookie at gmx.de (Max Lein) Date: Wed, 3 Sep 2014 08:20:19 -0400 Subject: [TxMt] Re: search & replace window borked in alpha.9561 In-Reply-To: References: Message-ID: <7C892C77-A975-4FD0-96FB-4CF538B410F1@gmx.de> On 03 Sep 2014, at 08:00, textmate-request at lists.macromates.com wrote: > I can?t reproduce this with 2.0-alpha.9563 but there have been a lot > of changes done to the find dialog, so I assume it has been fixed, > otherwise let me know. Shortly after my post, alpha.9563 was made available for download. Indeed, Search & Replace works as expected now, I think the issue has been fixed. Thanks! Max From wellsoliver at gmail.com Thu Sep 4 20:28:44 2014 From: wellsoliver at gmail.com (Wells Oliver) Date: Thu, 4 Sep 2014 13:28:44 -0700 Subject: [TxMt] Forgive me if this is a classic request, but can I have the full path? Message-ID: It'd be great if TextMate could show the full path in the title bar of the window, and in the Window drop-down because it's kind of driving me batty, having projects opened with similar file names. Is this an option? -- Wells Oliver wellsoliver at gmail.com -------------- next part -------------- An HTML attachment was scrubbed... URL: From graham.p.heath at gmail.com Thu Sep 4 20:34:29 2014 From: graham.p.heath at gmail.com (=?utf-8?Q?graham.p.heath=40gmail.com?=) Date: Thu, 4 Sep 2014 15:34:29 -0500 Subject: [TxMt] Re: Forgive me if this is a classic request, but can I have the full path? In-Reply-To: References: Message-ID: The file you're looking for is called .tm_properties, for reference see?https://gist.github.com/dvessel/1478685 It can be loaded from your home dir, or from the "project directory". Im using this blob, I forget exactly what it does, but basically, it shows the information of interest. windowTitle = "$windowTitle ${TM_DIRECTORY/\A(?:\/Users\/$USER\/Sites\/[\w-]+\/?(.*)|(.+))\z/${2:? ? ${2/\/Users\/$USER/~/}:${1/\A(?=.)/ ? /}}/}" Thanks --? graham.p.heath at gmail.com On September 4, 2014 at 3:29:06 PM, Wells Oliver (wellsoliver at gmail.com) wrote: It'd be great if TextMate could show the full path in the title bar of the window, and in the Window drop-down because it's kind of driving me batty, having projects opened with similar file names. Is this an option? -- Wells Oliver wellsoliver at gmail.com _______________________________________________ textmate mailing list textmate at lists.macromates.com http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From efrancos at efrancos.com Fri Sep 5 06:35:40 2014 From: efrancos at efrancos.com (Eduardo Francos) Date: Fri, 5 Sep 2014 08:35:40 +0200 Subject: [TxMt] Re: File changed status wrong when modified by a bundle command In-Reply-To: References: <20140809165408.534.qmail@f4mail-235-237.rediffmail.com> <08CDBA6E-DE1D-4F24-99E1-FCC7289FF120@textmate.org> <373CE520-7C57-4E46-A295-ED9C650F5EB5@efrancos.com> Message-ID: That was fast! Thanks :) On 2 sept. 2014, at 21:27, Allan Odgaard wrote: > On 17 Aug 2014, at 10:46, Eduardo Francos wrote: > >> As I said before the problem occurs only when the version file is modified by a command AND it's the currently active tab [?] > > Thanks for the test case, I have fixed the issue in 2.0-alpha.9563 (current nightly build). > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 495 bytes Desc: Message signed with OpenPGP using GPGMail URL: From efrancos at efrancos.com Fri Sep 5 07:55:33 2014 From: efrancos at efrancos.com (Eduardo Francos) Date: Fri, 5 Sep 2014 09:55:33 +0200 Subject: [TxMt] File status after a replace all Message-ID: Hi, I've found a document modified status issue when using "Replace all" with either project or folder Find in... dialogs When the Replace All button is clicked all the matches are applied to the documents containing them, but the status of the affected documents (which where not modified before issuing the command) does not change to "modified" as it could be expected. When you close the Find in... dialog it shows a dialog box with the option to save the affected documents. The text in the dialog states that the changes will be lost if you don't save them, which is actually incorrect, or at least misleading, because if you select not to save the documents they still include the applied changes. Their "modified" status though doesn't reflect this. To reproduce the bug just open two or more files that contain a common string, open a Find in... dialog and do a replace all of that string, check the documents' statuses. Unless I'm missing something obvious I believe that the additional "save documents" dialog is a lot of work just to save the "??S" I'd issue anyway in most cases, but not always. Also, given that the Find in... dialogs aren't modal, so the user can switch back to the documents and continue working, when he gets back to the Find dialog and closes it he may not always remember the context in which he executed the replace operation. When that happens my " keep the control" attitude always forces me to select not to save them ;-) If a builtin "save changes" option is nevertheless required then I believe that an additional "Replace All & Save" button could be more effective. To minimize real estate usage this button could be placed immediately to the right of the Replace All button with just "& Save" as the label. Alternatively, given that in this case the "Replace & Find" action button is apparently always disabled it could simply replace it. Thanks Gualo -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 495 bytes Desc: Message signed with OpenPGP using GPGMail URL: From matt at tidbits.com Fri Sep 5 14:15:09 2014 From: matt at tidbits.com (Matt Neuburg) Date: Fri, 5 Sep 2014 07:15:09 -0700 Subject: [TxMt] Re: File status after a replace all In-Reply-To: References: Message-ID: I agree with this; I'd put the problem another way, namely by describing how I use Replace All and what happens when I do. I do a Find All across multiple documents to see what's going to happen. I decide it's okay. I do a Replace All. Then I switch to my project and do a Save All just like I've always done. Now I come back to the dialog and try to start another find. And it says: Do you want to save the documents you just changed? Wait, I just did that! So there are two problems: (1) The Find dialog doesn't have a correct notion of whether the documents have been saved; and (2) It's none of the Find dialog's business anyway! Its job is to find and replace. That's all. The project is where we save. I'm going to switch to the project and do the save, just like I always do. That's pretty much the same as what Eduardo is saying. m. PS The way BBEdit solves this is: when you do a global find-and-replace across multiple documents, you get a secondary dialog _first_ asking whether you also want to save automatically. Thus the question is settled up front. I always want to save, so I wouldn't mind having to pass thru such a dialog. On Sep 5, 2014, at 12:55 AM, Eduardo Francos wrote: > Hi, > I've found a document modified status issue when using "Replace all" with either project or folder Find in... dialogs > > When the Replace All button is clicked all the matches are applied to the documents containing them, but the status of the affected documents (which where not modified before issuing the command) does not change to "modified" as it could be expected. > > When you close the Find in... dialog it shows a dialog box with the option to save the affected documents. > The text in the dialog states that the changes will be lost if you don't save them, which is actually incorrect, or at least misleading, because if you select not to save the documents they still include the applied changes. Their "modified" status though doesn't reflect this. > > To reproduce the bug just open two or more files that contain a common string, open a Find in... dialog and do a replace all of that string, check the documents' statuses. > > Unless I'm missing something obvious I believe that the additional "save documents" dialog is a lot of work just to save the "??S" I'd issue anyway in most cases, but not always. > Also, given that the Find in... dialogs aren't modal, so the user can switch back to the documents and continue working, when he gets back to the Find dialog and closes it he may not always remember the context in which he executed the replace operation. When that happens my " keep the control" attitude always forces me to select not to save them ;-) > If a builtin "save changes" option is nevertheless required then I believe that an additional "Replace All & Save" button could be more effective. > To minimize real estate usage this button could be placed immediately to the right of the Replace All button with just "& Save" as the label. Alternatively, given that in this case the "Replace & Find" action button is apparently always disabled it could simply replace it. > -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From mailinglist at textmate.org Sat Sep 6 08:08:46 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Sat, 06 Sep 2014 10:08:46 +0200 Subject: [TxMt] Re: File status after a replace all In-Reply-To: References: Message-ID: On 5 Sep 2014, at 9:55, Eduardo Francos wrote: > I've found a document modified status issue when using "Replace all" > with either project or folder Find in... dialogs [?] Besides the implementation issues (that I do intend to address and are one of the remaining issues that have kept me from removing the alpha label) there?s the user experience with the ?Do you want to save?? sheet (ignoring that dismissing changes doesn?t undo replacements done in open documents). Would people be OK with Replace All (after a folder search) auto-saving changes (for non-open documents)? So it would never ask about saving and for open documents, it would do the replacement without saving anything, but for documents that have no open tab, it would simply operate on the data on disk. Ideally after clicking Replace All the button would turn into Undo Replacements, which would give a way to revert the changes made to the non-open files. From efrancos at efrancos.com Sat Sep 6 10:40:47 2014 From: efrancos at efrancos.com (Eduardo Francos) Date: Sat, 6 Sep 2014 12:40:47 +0200 Subject: [TxMt] Re: File status after a replace all In-Reply-To: References: Message-ID: <3CBBD9AC-0E91-4E48-9BAB-C66ED4743335@efrancos.com> On 6 sept. 2014, at 10:08, Allan Odgaard wrote: > On 5 Sep 2014, at 9:55, Eduardo Francos wrote: > >> I've found a document modified status issue when using "Replace all" with either project or folder Find in... dialogs [?] > Would people be OK with Replace All (after a folder search) auto-saving changes (for non-open documents)? So it would never ask about saving and for open documents, it would do the replacement without saving anything, but for documents that have no open tab, it would simply operate on the data on disk. That seems a good solution to me. Maybe it should handle opened but otherwise unmodified documents the same way as closed documents, and automatically save them as well. It should also be possible to add a .tm_properties setting to have all the files saved including those with an active editor window. Some may prefer to work that way. > Ideally after clicking Replace All the button would turn into Undo Replacements, which would give a way to revert the changes made to the non-open files. An Undo Replacements could require making the Find in... dialog modal, otherwise the user may edit some files before closing the Find dialog, possibly rendering impossible a full and consistent undo operation. Alternatively, when the user modifies any document then the "Undo Replacements" button could be disabled. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 495 bytes Desc: Message signed with OpenPGP using GPGMail URL: From matt at tidbits.com Sat Sep 6 14:15:12 2014 From: matt at tidbits.com (Matt Neuburg) Date: Sat, 6 Sep 2014 07:15:12 -0700 Subject: [TxMt] Re: File status after a replace all In-Reply-To: References: Message-ID: <6949BEFD-0013-4979-9E63-761CA5F3CEB3@tidbits.com> On Sep 6, 2014, at 1:08 AM, Allan Odgaard wrote: > Would people be OK with Replace All (after a folder search) auto-saving changes (for non-open documents)? So it would never ask about saving and for open documents, it would do the replacement without saving anything, but for documents that have no open tab, it would simply operate on the data on disk. I'm someone who does a _lot_ of folder-wide find-and-replace (I still think of this as project-wide, sorry) and this sounds great to me! The need to save manually afterwards is something I've gotten used to but I've always been somewhat mystified by it. Basically I always want to find/replace/save. (After all, I'm always under version control, and if I'm in any doubt of the wisdom of a particular change in a particular place, even after previewing what's going to happen, I can always open that document and run through it manually and separately. In the case of very complex and scary regular expression global-find-and-replace I have been known to open every single document individually and do the find-and-replace manually over each one, and I don't mind that at all.) m. PS This is not directly relevant, but the manual Find dialog has a Replace & Find button but lacks a Replace button (i.e. replace but don't find, so I can inspect the change I just made). It desperately needs this. A common pattern is Replace (but don't Find), look at what this did, now do a Next, then repeat, so that you can look at what you're doing as you go. This is so important that I often switch to BBEdit just to do my replacing! And BBEdit has keyboard shortcuts for both of these. Indeed, in my opinion Replace (but don't Find) should be the default; Replace & Find should be a separate option, possibly its alternate when you hold the Option key or similar. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From matt at tidbits.com Sat Sep 6 14:21:51 2014 From: matt at tidbits.com (Matt Neuburg) Date: Sat, 6 Sep 2014 07:21:51 -0700 Subject: [TxMt] Re: File status after a replace all In-Reply-To: References: Message-ID: <234FB365-6F8D-4141-8867-4F2CF2CC873C@tidbits.com> Sorry to keep switching gears, but TM now does a thing where Command-G takes me into the next document instead of wrapping around in this document. Is there a way I can get it to stop doing that? By the time I get to the command-G I'm inspecting the individual documents. m. > PS This is not directly relevant, but the manual Find dialog has a Replace & Find button but lacks a Replace button (i.e. replace but don't find, so I can inspect the change I just made). It desperately needs this. A common pattern is Replace (but don't Find), look at what this did, now do a Next, then repeat, so that you can look at what you're doing as you go. This is so important that I often switch to BBEdit just to do my replacing! And BBEdit has keyboard shortcuts for both of these. Indeed, in my opinion Replace (but don't Find) should be the default; Replace & Find should be a separate option, possibly its alternate when you hold the Option key or similar. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From mylists at ct-jung.de Sat Sep 6 15:48:30 2014 From: mylists at ct-jung.de (Tobias Jung) Date: Sat, 6 Sep 2014 17:48:30 +0200 Subject: [TxMt] Symbol list sorting Message-ID: <20140906154830.157728978@w010a4a1.kasserver.com> Hello, in TM 1 the symbol list (on the right-hand side of the status bar) was sorted alphabetically, is there some way to get this back in TM 2? Kind regards, Tobias Jung From matt at tidbits.com Sun Sep 7 01:49:37 2014 From: matt at tidbits.com (Matt Neuburg) Date: Sat, 6 Sep 2014 18:49:37 -0700 Subject: [TxMt] hang (sort of) when we right-arrow over a right-curly-brace Message-ID: <2A807F41-13E4-4B0D-A49B-D693C322DC3A@tidbits.com> If I'm repeatedly typing right-arrow to move the cursor position to the right, and the character I pass over happens to be a right-curly-brace that isn't balanced by a nearby left-curly-brace, TM seems to hang. It does recover, but it scares the pants off me every time it happens. m. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From mailinglist at textmate.org Sun Sep 7 09:44:06 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Sun, 07 Sep 2014 11:44:06 +0200 Subject: [TxMt] Re: hang (sort of) when we right-arrow over a right-curly-brace In-Reply-To: <2A807F41-13E4-4B0D-A49B-D693C322DC3A@tidbits.com> References: <2A807F41-13E4-4B0D-A49B-D693C322DC3A@tidbits.com> Message-ID: On 7 Sep 2014, at 3:49, Matt Neuburg wrote: > If I'm repeatedly typing right-arrow to move the cursor position to > the right, and the character I pass over happens to be a > right-curly-brace that isn't balanced by a nearby left-curly-brace, TM > seems to hang. It does recover, but it scares the pants off me every > time it happens. As mentioned previously?, the function is slower than it needs to be because it always uses regular expressions for the search, improving it is on the to-do. ? http://lists.macromates.com/textmate/2014-April/037237.html From mailinglist at textmate.org Sun Sep 7 09:45:52 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Sun, 07 Sep 2014 11:45:52 +0200 Subject: [TxMt] Re: Symbol list sorting In-Reply-To: <20140906154830.157728978@w010a4a1.kasserver.com> References: <20140906154830.157728978@w010a4a1.kasserver.com> Message-ID: <504286BA-0396-420D-9874-49430E657A19@textmate.org> On 6 Sep 2014, at 17:48, Tobias Jung wrote: > in TM 1 the symbol list (on the right-hand side of the status bar) was > sorted alphabetically, is there some way to get this back in TM 2? Sorting the symbols is not supported in 2.0. In what sort of document types would you prefer the list sorted? And if you are unaware, using ??T opens a window that supports ?type to filter?. From timothy.c.bates at gmail.com Sun Sep 7 10:07:25 2014 From: timothy.c.bates at gmail.com (Timothy Bates) Date: Sun, 07 Sep 2014 03:07:25 -0700 (PDT) Subject: [TxMt] Re: Symbol list sorting In-Reply-To: <504286BA-0396-420D-9874-49430E657A19@textmate.org> References: <504286BA-0396-420D-9874-49430E657A19@textmate.org> Message-ID: <1410084444828.01e113f6@Nodemailer> ??T ?always beats the symbol list in terms of speed? It rocks! The symbol list is nice, however, to show which symbol one is inside. ? Sent from Mailbox On Sun, Sep 7, 2014 at 10:46 AM, Allan Odgaard wrote: > On 6 Sep 2014, at 17:48, Tobias Jung wrote: >> in TM 1 the symbol list (on the right-hand side of the status bar) was >> sorted alphabetically, is there some way to get this back in TM 2? > Sorting the symbols is not supported in 2.0. > In what sort of document types would you prefer the list sorted? > And if you are unaware, using ??T opens a window that supports > ?type to filter?. > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From mailinglist at textmate.org Sun Sep 7 10:08:50 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Sun, 07 Sep 2014 12:08:50 +0200 Subject: [TxMt] Re: File status after a replace all In-Reply-To: <234FB365-6F8D-4141-8867-4F2CF2CC873C@tidbits.com> References: <234FB365-6F8D-4141-8867-4F2CF2CC873C@tidbits.com> Message-ID: On 6 Sep 2014, at 16:21, Matt Neuburg wrote: > Sorry to keep switching gears, but TM now does a thing where Command-G > takes me into the next document instead of wrapping around in this > document. Is there a way I can get it to stop doing that? It always did this ? you can ?break? it by pressing ?E (i.e. place selection on the search clipboard, since the selection is a match, so it should go to next match on ?G, but it?ll drop the associated next document). It won?t work if you did a regexp search. I can add a (hidden) setting if it?s a general issue (and people would prefer to never have it continue to next document). >> PS This is not directly relevant, but the manual Find dialog has a >> Replace & Find button but lacks a Replace button [?] in my opinion >> Replace (but don't Find) should be the default; Replace & Find should >> be a separate option, possibly its alternate when you hold the Option >> key or similar. For a non-regexp search I generally use paste (?V) for ?replace? ? though the main reason for lack of a replace button is mainky the width of 6 buttons in the dialog making it rather wide. But I could probably fold the two replace buttons into one, as you suggest. Was also thinking about folding all three replace actions into an action menu button (like the print dialog?s PDF button), but I think people would dislike having to go through a menu when doing multiple replacements from the UI. As for key equivalents (for Replace), if it needs to contain G then we only have ??G and ???G left. From mylists at ct-jung.de Sun Sep 7 13:20:22 2014 From: mylists at ct-jung.de (Tobias Jung) Date: Sun, 7 Sep 2014 15:20:22 +0200 Subject: [TxMt] Re: Symbol list sorting In-Reply-To: <504286BA-0396-420D-9874-49430E657A19@textmate.org> References: <20140906154830.157728978@w010a4a1.kasserver.com> <504286BA-0396-420D-9874-49430E657A19@textmate.org> Message-ID: <20140907132022.1013859497@w010a4a1.kasserver.com> Allan Odgaard wrote (Sun, 7 Sep 2014 11:45:52 +0200): > On 6 Sep 2014, at 17:48, Tobias Jung wrote: > >> in TM 1 the symbol list (on the right-hand side of the status bar) was >> sorted alphabetically, is there some way to get this back in TM 2? > > Sorting the symbols is not supported in 2.0. > > In what sort of document types would you prefer the list sorted? In PHP files to jump to a certain function/method. I usually remember the name (e.g. getSingleArticle) so in TM1, I used to open the symbol list and found it quickly because of the sorting. > And if you are unaware, using ??T opens a window that supports > ?type to filter?. Uuuh, I have to admit that I wasn't aware of that. I just tried it, and it certainly does the trick -- and it's even faster for me because my hands don't even have to leave the keyboard! Thanks a lot! :-) Kind regards, Tobias Jung From matt at tidbits.com Sun Sep 7 15:45:48 2014 From: matt at tidbits.com (Matt Neuburg) Date: Sun, 7 Sep 2014 08:45:48 -0700 Subject: [TxMt] Re: hang (sort of) when we right-arrow over a right-curly-brace In-Reply-To: References: <2A807F41-13E4-4B0D-A49B-D693C322DC3A@tidbits.com> Message-ID: Wasn't trying to yank your chain, sorry; glad this is known. One thing that might help here is some intelligent scoping; there's no need to look back more than a dozen characters, because the scope in which the curly braces are meaningful comes to an end at that point. Not sure how TM could be told about that... m. On Sep 7, 2014, at 2:44 AM, Allan Odgaard wrote: > On 7 Sep 2014, at 3:49, Matt Neuburg wrote: > >> If I'm repeatedly typing right-arrow to move the cursor position to the right, and the character I pass over happens to be a right-curly-brace that isn't balanced by a nearby left-curly-brace, TM seems to hang. It does recover, but it scares the pants off me every time it happens. > > As mentioned previously?, the function is slower than it needs to be because it always uses regular expressions for the search, improving it is on the to-do. > > ? http://lists.macromates.com/textmate/2014-April/037237.html > -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From matt at tidbits.com Sun Sep 7 15:46:45 2014 From: matt at tidbits.com (Matt Neuburg) Date: Sun, 7 Sep 2014 08:46:45 -0700 Subject: [TxMt] Re: Symbol list sorting In-Reply-To: <504286BA-0396-420D-9874-49430E657A19@textmate.org> References: <20140906154830.157728978@w010a4a1.kasserver.com> <504286BA-0396-420D-9874-49430E657A19@textmate.org> Message-ID: <9C5696B5-E920-48C0-BCDF-3B6C7D81A860@tidbits.com> On Sep 7, 2014, at 2:45 AM, Allan Odgaard wrote: > And if you are unaware, using ??T opens a window that supports ?type to filter?. This is definitely the Tip of the Day! m. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From carpii.uk at gmail.com Mon Sep 8 23:36:26 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Tue, 9 Sep 2014 00:36:26 +0100 Subject: [TxMt] Re: hang (sort of) when we right-arrow over a right-curly-brace In-Reply-To: References: <2A807F41-13E4-4B0D-A49B-D693C322DC3A@tidbits.com> Message-ID: On 7 September 2014 16:45, Matt Neuburg wrote: > the scope in which the curly braces are meaningful comes to an end at > that point. > Hmm, I don't agree with that. For one line of code perhaps, but its useful for nested blocks too. There are also cases where you have code collapsed, and are trying to balance the braces, in these cases there's just no sensible limit you can apply. Perhaps there should be an option to disable the automatic highlighting, and then a key combo which triggers it manually for those cases when you need it? -------------- next part -------------- An HTML attachment was scrubbed... URL: From carpii.uk at gmail.com Mon Sep 8 23:45:54 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Tue, 9 Sep 2014 00:45:54 +0100 Subject: [TxMt] Any solutions for file notifications on remote fs? Message-ID: I have a recurring problem where TM cannot detect file changes. This is because I work on a a remote filesystem (Centos), which is mounted via AFP. Its actually a VM running on ESXI Server on my LAN, although a Vagrant or VMWare setup is quite common now. >From what I understand, TM uses kqueue or BSD's FSEvents for file notifications (I suspect kqueue?). Its causing me a lot of problems as I switch git branch and TM doesn't always detect it, sometimes resulting in a file from another branch being saved over the top. A bit of googling shows there is libkqueue, which tries to provide a kqueue interface on top of the underlying host OS, but its not clear to me whether this would be suitable for a remote fs, even if it was feasible to compile into TM. Has anyone found a workaround to this please? -------------- next part -------------- An HTML attachment was scrubbed... URL: From matt at tidbits.com Tue Sep 9 02:32:48 2014 From: matt at tidbits.com (Matt Neuburg) Date: Mon, 8 Sep 2014 19:32:48 -0700 Subject: [TxMt] Re: hang (sort of) when we right-arrow over a right-curly-brace In-Reply-To: References: <2A807F41-13E4-4B0D-A49B-D693C322DC3A@tidbits.com> Message-ID: <353B0964-9048-4C69-812A-77C508F6C883@tidbits.com> On Sep 8, 2014, at 4:36 PM, Carpii UK wrote: > Hmm, I don't agree with that. > > For one line of code perhaps, but its useful for nested blocks too. > Yes, but your are missing my point, which is that I'm not in that situation. My AsciiDoc document looks like this: [Lots and lots and lots of text]`howdy}` As we right-arrow over the right curly brace, there is no need to look back further than "howdy"; after that, we're not in the same curly-brace-matching scope anyway. I'm not saying that this heuristic is feasible everywhere - just that it's perfectly obvious to a human what the situation is at a glance ("of *course* it's unbalanced!"), so it seems unnecessary for TextMate to take so long to make up its mind. m. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From mailinglist at textmate.org Tue Sep 9 03:59:18 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 09 Sep 2014 05:59:18 +0200 Subject: [TxMt] Re: Any solutions for file notifications on remote fs? In-Reply-To: References: Message-ID: On 9 Sep 2014, at 1:45, Carpii UK wrote: > This is because I work on a a remote filesystem (Centos), which is > mounted > via AFP. > Its actually a VM running on ESXI Server on my LAN, although a Vagrant > or > VMWare setup is quite common now. The typical Vagrant setup will mount your Mac (project) folder in the virtual machine. This means on the Mac it?s a 100% local file system with all the advantages that this brings, yet the files are still instantly accessible from the virtual machine. > [?] A bit of googling shows there is libkqueue, which tries to > provide a kqueue > interface on top of the underlying host OS, but its not clear to me > whether > this would be suitable for a remote fs Either the remote machine will inform the client about file systme changes (mounting a Mac via AFP does support this) or alternatively the client will have to poll the remote server to test if there are changes (as would be required in your case). I would be surprised if libkqueue includes code to do regular polling. > Has anyone found a workaround to this please? As indicated above, a workaround is to have the virtual machine mount a folder from your Mac. From ken+textmate at ken.net Sat Sep 13 13:11:30 2014 From: ken+textmate at ken.net (Ken Snyder) Date: Sat, 13 Sep 2014 14:11:30 +0100 Subject: [TxMt] Dockerfile Bundle Message-ID: Is it possible to offer the Dockerfile bundle as part of the TextMate config options? I think there's enough popularity (at least momentum) behind Docker to make this worthy of being listed on the master list. https://github.com/docker/docker/tree/master/contrib/syntax/textmate Ken -------------- next part -------------- An HTML attachment was scrubbed... URL: From mummer at whitefalls.org Tue Sep 16 09:52:35 2014 From: mummer at whitefalls.org (Michael Sheets) Date: Tue, 16 Sep 2014 04:52:35 -0500 Subject: [TxMt] Re: Dockerfile Bundle In-Reply-To: References: Message-ID: On Sep 13, 2014, at 8:11 AM, Ken Snyder wrote: > Is it possible to offer the Dockerfile bundle as part of the TextMate config options? I think there's enough popularity (at least momentum) behind Docker to make this worthy of being listed on the master list. > > https://github.com/docker/docker/tree/master/contrib/syntax/textmate I have already replied to Ken off-list but for others wanting to get a bundle added to the installer in 2.0 email me at: bundles(at)textmate.org From mailinglist at textmate.org Tue Sep 16 10:27:51 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 16 Sep 2014 12:27:51 +0200 Subject: [TxMt] Re: Dockerfile Bundle In-Reply-To: References: Message-ID: <20B55CE8-7126-4848-8F1E-32FAB4C8AB52@textmate.org> On 16 Sep 2014, at 11:52, Michael Sheets wrote: > I have already replied to Ken off-list but for others wanting to get a > bundle added to the installer [?] It?s also worth mentioning that in 2.0 we have ?project scopes? for identifyable ?projects?. The current list is hardcoded in TextMate here: https://github.com/textmate/textmate/blob/master/Frameworks/DocumentWindow/src/DocumentController.mm#L1211 It?s a list I?ll gladly extend. The advantage is that it allow bundle items (e.g. with key bindings) to be scoped to specific ?project types?, which is how we have all the version control bundles share the ?Y key or the various build systems use ?B, even though the current file type is not sufficient to figure out version control or build system. I believe I mentioned it before, but long-term this should also allow us to suggest bundles to install when TextMate recognizes what the user is working with, as they may not bother to look for a bundle for the thing themselves. From me at igorkozlov.me Tue Sep 16 12:55:45 2014 From: me at igorkozlov.me (Igor K) Date: Tue, 16 Sep 2014 15:55:45 +0300 Subject: [TxMt] Yosemite view effects Message-ID: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> Looks like Yosemite exposes NSVisualEffectView class for those fancy translucency and blur effects https://developer.apple.com/library/prerelease/mac/documentation/Foundation/Reference/NSVisualEffectView_Class/index.html I understand that I probably rush things up a bit, but are there any plans to support those in near future? -------------- next part -------------- An HTML attachment was scrubbed... URL: From caiofbpa at icloud.com Tue Sep 16 13:00:51 2014 From: caiofbpa at icloud.com (Caio Fernando Bertoldi Paes de Andrade) Date: Tue, 16 Sep 2014 10:00:51 -0300 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> Message-ID: <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> On a side note, the tabs on Yosemite look odd, it would be great if TextMate could use the default system tabs construction there. Caio > On 16 of Sep, 2014, at 09:55, Igor K wrote: > > Looks like Yosemite exposes NSVisualEffectView class for those fancy translucency and blur effects https://developer.apple.com/library/prerelease/mac/documentation/Foundation/Reference/NSVisualEffectView_Class/index.html > > I understand that I probably rush things up a bit, but are there any plans to support those in near future? > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From me at igorkozlov.me Tue Sep 16 13:04:32 2014 From: me at igorkozlov.me (Igor K) Date: Tue, 16 Sep 2014 16:04:32 +0300 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> Message-ID: <3970B09E-4404-4C6C-84AE-5E8012410325@igorkozlov.me> Can?t agree more. Tab bar like in Terminal.app for example will look really neat and native. > On 16 09 2014, at 16:00, Caio Fernando Bertoldi Paes de Andrade wrote: > > On a side note, the tabs on Yosemite look odd, it would be great if TextMate could use the default system tabs construction there. > > Caio > >> On 16 of Sep, 2014, at 09:55, Igor K > wrote: >> >> Looks like Yosemite exposes NSVisualEffectView class for those fancy translucency and blur effects https://developer.apple.com/library/prerelease/mac/documentation/Foundation/Reference/NSVisualEffectView_Class/index.html >> >> I understand that I probably rush things up a bit, but are there any plans to support those in near future? >> >> _______________________________________________ >> textmate mailing list >> textmate at lists.macromates.com >> http://lists.macromates.com/listinfo/textmate > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From mailinglist at textmate.org Tue Sep 16 13:28:33 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 16 Sep 2014 15:28:33 +0200 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> Message-ID: <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> On 16 Sep 2014, at 15:00, Caio Fernando Bertoldi Paes de Andrade wrote: >> On 16 of Sep, 2014, at 09:55, Igor K wrote: >> >> [?] I understand that I probably rush things up a bit, but are >> there any plans to support those in near future? No plans to support them in near future. Adding support for this would be cumbersome without dropping backwards compatibility and I am not sure where you even want us to add these views. > On a side note, the tabs on Yosemite look odd, it would be great if > TextMate could use the default system tabs construction there. There are no ?default system tabs construction?. I?ll probably revise a few things for Yosemite (UI-wise) but I?m leaning toward first dropping backwards compatibility, as I think getting the proper look is not easily done in a backwards compatibile way. From me at igorkozlov.me Tue Sep 16 13:33:31 2014 From: me at igorkozlov.me (Igor K) Date: Tue, 16 Sep 2014 16:33:31 +0300 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> Message-ID: > On 16 09 2014, at 16:28, Allan Odgaard wrote: > > I am not sure where you even want us to add these views. The main editor view of course. Right behind the text like in Terminal.app?s pre-installed Homebrew theme. -------------- next part -------------- An HTML attachment was scrubbed... URL: From caiofbpa at icloud.com Tue Sep 16 14:05:01 2014 From: caiofbpa at icloud.com (Caio Fernando Bertoldi Paes de Andrade) Date: Tue, 16 Sep 2014 11:05:01 -0300 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> Message-ID: <7903873F-9D3D-4EB5-A45A-C97E88E368C4@icloud.com> >> I am not sure where you even want us to add these views. > > The main editor view of course. Right behind the text like in Terminal.app?s pre-installed Homebrew theme. Allan is right, it?s tricky to choose where to put those. I thought of having transparency only in the File Browser, like a sidebar app, but you had another idea. It?s not a simple unanimous design decision. >> There are no ?default system tabs construction?. I thought that NSTabView would be used by Apple apps to make those tabs, but I was clearly wrong. Well, I guess we?ll have to wait a bit more on this, which is perfectly fine. :-) Caio -------------- next part -------------- An HTML attachment was scrubbed... URL: From mailinglist at textmate.org Tue Sep 16 14:19:05 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 16 Sep 2014 16:19:05 +0200 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <7903873F-9D3D-4EB5-A45A-C97E88E368C4@icloud.com> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> <7903873F-9D3D-4EB5-A45A-C97E88E368C4@icloud.com> Message-ID: <35A7E526-74A4-4736-9FBE-7D64BEEA7EE6@textmate.org> On 16 Sep 2014, at 16:05, Caio Fernando Bertoldi Paes de Andrade wrote: > [?] I thought of having transparency only in the File Browser, like > a sidebar app This is what Apple is suggesting, though only for source lists. TextMate?s file browser isn?t a source list, but for people who prefer this look, it can be made to use the style by running: defaults write com.macromates.TextMate.preview fileBrowserStyle SourceList I believe that should give a blurred transparent background on Yosemite. From me at igorkozlov.me Tue Sep 16 14:27:55 2014 From: me at igorkozlov.me (Igor K) Date: Tue, 16 Sep 2014 17:27:55 +0300 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <35A7E526-74A4-4736-9FBE-7D64BEEA7EE6@textmate.org> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> <7903873F-9D3D-4EB5-A45A-C97E88E368C4@icloud.com> <35A7E526-74A4-4736-9FBE-7D64BEEA7EE6@textmate.org> Message-ID: > On 16 09 2014, at 17:19, Allan Odgaard wrote: > > I believe that should give a blurred transparent background on Yosemite. It does work, thanks. And looks like nothing changes but the visual representation and that?s a good thing. Still I think that native TM support using a ?defaults? option for the editor?s view blur would be a real nice feature. -------------- next part -------------- An HTML attachment was scrubbed... URL: From doob at me.com Tue Sep 16 14:31:21 2014 From: doob at me.com (Jacob Carlborg) Date: Tue, 16 Sep 2014 16:31:21 +0200 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <7903873F-9D3D-4EB5-A45A-C97E88E368C4@icloud.com> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> <7903873F-9D3D-4EB5-A45A-C97E88E368C4@icloud.com> Message-ID: On 16/09/14 16:05, Caio Fernando Bertoldi Paes de Andrade wrote: > Allan is right, it?s tricky to choose where to put those. I thought of > having transparency only in the File Browser, like a sidebar app, but > you had another idea. It?s not a simple unanimous design decision. If I recall correctly, table views configured as a source list will get the vibrant appearance automatically. You can set the file browser to be displayed as a source list with the hidden setting "fileBrowserStyle", see [1]. [1] https://github.com/textmate/textmate/wiki/Hidden-Settings -- /Jacob Carlborg From doob at me.com Tue Sep 16 14:45:03 2014 From: doob at me.com (Jacob Carlborg) Date: Tue, 16 Sep 2014 16:45:03 +0200 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> Message-ID: On 16/09/14 15:28, Allan Odgaard wrote: > There are no ?default system tabs construction?. When we're talking about the style of tabs. It would be nice if TextMate could use MMTabBarView [1][2] for its tab implementation. Although it don't think it supports the Yosemite style but it do support the style used in 10.9 in Safari and Finder. [1] https://github.com/mimo42/MMTabBarView/ [2] http://mimo42.github.io/MMTabBarView/ -- /Jacob Carlborg From efrancos at efrancos.com Wed Sep 17 11:11:21 2014 From: efrancos at efrancos.com (Eduardo Francos) Date: Wed, 17 Sep 2014 13:11:21 +0200 Subject: [TxMt] Bug in enclose selection with multi-key characters using US-International keyboard layout Message-ID: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> Hi, When using the US-International keyboard layout, if you want to enclose selected text with quotes, single quotes or back-quotes by typing them, the originally selected text is deleted. Those keys are used for diacritical accents by the US-International layout and require two keystrokes, the quote character itself followed by either a space to indicate you actually want the quote character itself, or a different character if you want to apply the diacritical accent on it. For instance a single quote ' followed by an a gives ?. If the quote character is followed by a character that doesn't accept diacritical accents then the layout issues the two characters as they were typed, so a " followed by a w gives "w I do most of my programming using the standard US layout but sometimes I need to switch to spanish or french. A really insignificant bug, I know :p and I can certainly live with it, so this is mostly for information. It's good to know someone knows it's there :-) Gualo -------------- next part -------------- An HTML attachment was scrubbed... URL: -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 495 bytes Desc: Message signed with OpenPGP using GPGMail URL: From peter at boctor.net Wed Sep 17 19:50:41 2014 From: peter at boctor.net (Peter Boctor) Date: Wed, 17 Sep 2014 12:50:41 -0700 Subject: [TxMt] Yosemite 'Go To File' delay Message-ID: <36F88858-8BBE-4B91-8F25-592BE274F185@boctor.net> I just installed the Yosemite preview and now there's a noticeable delay with I hit Cmd-T. Now when I hit Cmd-T followed by return the app beeps, because the Go To File window hasn't fully loaded yet. I have to hit Cmd-T, pause then hit return. This is on the same project, so nothing else has changed. Anyone else seeing this too? Peter From timothy.c.bates at gmail.com Wed Sep 17 19:52:41 2014 From: timothy.c.bates at gmail.com (Timothy Bates) Date: Wed, 17 Sep 2014 12:52:41 -0700 (PDT) Subject: [TxMt] Re: Yosemite 'Go To File' delay In-Reply-To: <36F88858-8BBE-4B91-8F25-592BE274F185@boctor.net> References: <36F88858-8BBE-4B91-8F25-592BE274F185@boctor.net> Message-ID: <1410983561004.314a7c21@Nodemailer> also for my new favorite cntrl option cmd V ? Sent from Mailbox On Wed, Sep 17, 2014 at 8:50 PM, Peter Boctor wrote: > I just installed the Yosemite preview and now there's a noticeable delay with I hit Cmd-T. > Now when I hit Cmd-T followed by return the app beeps, because the Go To File window hasn't fully loaded yet. I have to hit Cmd-T, pause then hit return. > This is on the same project, so nothing else has changed. > Anyone else seeing this too? > Peter > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From gweltaz at ucsc.edu Wed Sep 17 19:57:04 2014 From: gweltaz at ucsc.edu (Gildas Hamel) Date: Wed, 17 Sep 2014 12:57:04 -0700 Subject: [TxMt] Re: Yosemite 'Go To File' delay In-Reply-To: <1410983561004.314a7c21@Nodemailer> References: <36F88858-8BBE-4B91-8F25-592BE274F185@boctor.net> <1410983561004.314a7c21@Nodemailer> Message-ID: <20140917195704.GA40305@eduroam-233-151.ucsc.edu> * Skriv a reas Timothy Bates (timothy.c.bates at gmail.com): |> also for my new favorite cntrl option cmd V |> |> On Wed, Sep 17, 2014 at 8:50 PM, Peter Boctor wrote: |> |> > I just installed the Yosemite preview and now there's a noticeable delay with I hit Cmd-T. |> > Now when I hit Cmd-T followed by return the app beeps, because the Go To File window hasn't fully loaded yet. I have to hit Cmd-T, pause then hit return. |> > This is on the same project, so nothing else has changed. |> > Anyone else seeing this too? I don't see either of the issues. I'm on Yosemite beta 3. Actually, I am under the *impression* everything is faster. --Gildas From peter at boctor.net Wed Sep 17 20:00:58 2014 From: peter at boctor.net (Peter Boctor) Date: Wed, 17 Sep 2014 13:00:58 -0700 Subject: [TxMt] Re: Yosemite 'Go To File' delay In-Reply-To: <20140917195704.GA40305@eduroam-233-151.ucsc.edu> References: <36F88858-8BBE-4B91-8F25-592BE274F185@boctor.net> <1410983561004.314a7c21@Nodemailer> <20140917195704.GA40305@eduroam-233-151.ucsc.edu> Message-ID: <0359777B-A0BB-4A21-86A9-0285DFF15511@boctor.net> FWIW, I'm on Developer Preview 8 (14A361c). Peter > On Sep 17, 2014, at 12:57 PM, Gildas Hamel wrote: > > * Skriv a reas Timothy Bates (timothy.c.bates at gmail.com): > |> also for my new favorite cntrl option cmd V > |> > |> On Wed, Sep 17, 2014 at 8:50 PM, Peter Boctor wrote: > |> > |> > I just installed the Yosemite preview and now there's a noticeable delay with I hit Cmd-T. > |> > Now when I hit Cmd-T followed by return the app beeps, because the Go To File window hasn't fully loaded yet. I have to hit Cmd-T, pause then hit return. > |> > This is on the same project, so nothing else has changed. > |> > Anyone else seeing this too? > > I don't see either of the issues. I'm on Yosemite beta 3. Actually, I am under the *impression* everything is faster. > --Gildas From caiofbpa at icloud.com Wed Sep 17 20:45:14 2014 From: caiofbpa at icloud.com (Caio Fernando Bertoldi Paes de Andrade) Date: Wed, 17 Sep 2014 17:45:14 -0300 Subject: [TxMt] Re: Yosemite 'Go To File' delay In-Reply-To: <0359777B-A0BB-4A21-86A9-0285DFF15511@boctor.net> References: <36F88858-8BBE-4B91-8F25-592BE274F185@boctor.net> <1410983561004.314a7c21@Nodemailer> <20140917195704.GA40305@eduroam-233-151.ucsc.edu> <0359777B-A0BB-4A21-86A9-0285DFF15511@boctor.net> Message-ID: <99A4A73F-2FBD-495B-B1FA-C486D93C69EA@icloud.com> I?m on DP8 and I notice a very small delay on Cmd+T, I can reproduce the beep you talked about if I hit enter very quickly, but that isn?t enough to bother me. Caio > On 17 of Sep, 2014, at 17:00, Peter Boctor wrote: > > FWIW, I'm on Developer Preview 8 (14A361c). > > Peter > >> On Sep 17, 2014, at 12:57 PM, Gildas Hamel wrote: >> >> * Skriv a reas Timothy Bates (timothy.c.bates at gmail.com): >> |> also for my new favorite cntrl option cmd V >> |> >> |> On Wed, Sep 17, 2014 at 8:50 PM, Peter Boctor wrote: >> |> >> |> > I just installed the Yosemite preview and now there's a noticeable delay with I hit Cmd-T. >> |> > Now when I hit Cmd-T followed by return the app beeps, because the Go To File window hasn't fully loaded yet. I have to hit Cmd-T, pause then hit return. >> |> > This is on the same project, so nothing else has changed. >> |> > Anyone else seeing this too? >> >> I don't see either of the issues. I'm on Yosemite beta 3. Actually, I am under the *impression* everything is faster. >> --Gildas > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate From mailinglist at textmate.org Thu Sep 18 07:48:04 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Thu, 18 Sep 2014 09:48:04 +0200 Subject: [TxMt] Re: Bug in enclose selection with multi-key characters using US-International keyboard layout In-Reply-To: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> References: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> Message-ID: <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> On 17 Sep 2014, at 13:11, Eduardo Francos wrote: > When using the US-International keyboard layout, if you want to > enclose selected text with quotes, single quotes or back-quotes by > typing them, the originally selected text is deleted. I have both the ISO and ANSI US/international keyboard, and I can type all of the following quotes without using dead keys: ???, ???, "?", '?', and `?`. > A really insignificant bug, I know :p and I can certainly live with > it, so this is mostly for information. It's good to know someone knows > it's there :-) I fear it would be rather complex to do a workaround for this because we would need to delay overtyping a selection when we detect the user starts a complex input sequence, and for most users I imagine this would be undesired (not seeing the selection immediately overtyped). I think it?d be fair to say that characters used for auto-pairing have the same limitations as characters available for key equivalents, where your issue also exist, i.e. you can?t bind a menu item to characters that you cannot type with a single key sans modifiers. From mailinglist at textmate.org Thu Sep 18 07:51:20 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Thu, 18 Sep 2014 09:51:20 +0200 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> Message-ID: <7C268632-2F53-407F-A521-7A87A2EA4311@textmate.org> On 16 Sep 2014, at 16:45, Jacob Carlborg wrote: > On 16/09/14 15:28, Allan Odgaard wrote: > >> There are no ?default system tabs construction?. > > When we're talking about the style of tabs. It would be nice if > TextMate could use MMTabBarView [1][2] for its tab implementation. > Although it don't think it supports the Yosemite style but it do > support the style used in 10.9 in Safari and Finder. Is there any reason you want TextMate to use this tab control? The current code already has something close to the 10.9 look and going forward, we want to get the Yosemite look, which, as you point out, the mentioned control does not (yet?) support. From doob at me.com Thu Sep 18 08:04:10 2014 From: doob at me.com (Jacob Carlborg) Date: Thu, 18 Sep 2014 10:04:10 +0200 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <7C268632-2F53-407F-A521-7A87A2EA4311@textmate.org> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> <7C268632-2F53-407F-A521-7A87A2EA4311@textmate.org> Message-ID: On 18/09/14 09:51, Allan Odgaard wrote: > Is there any reason you want TextMate to use this tab control? The reason is because that tab control has a look that closer matches the look in Safari and Finder. The look TextMate uses is something more close to the look in Terminal. I just thinks the one used by Safari is nicer looking. I also got the idea that TextMate is using PSMTabBarControl, which MMTabBarView claims to be fairly compatible with. So I was hoping it would be an easy change. Although I can be completely wrong. > The current code already has something close to the 10.9 look and going > forward, we want to get the Yosemite look, which, as you point out, the > mentioned control does not (yet?) support. This suggestion might have come a bit late, I don't know for how long you want to support the 10.9 look. I hope as long as 10.9 is supported. Compared to iTerm which has the Yosemite look on 10.9, which looks horrible. -- /Jacob Carlborg From doob at me.com Thu Sep 18 08:10:40 2014 From: doob at me.com (Jacob Carlborg) Date: Thu, 18 Sep 2014 10:10:40 +0200 Subject: [TxMt] Closing the new commit window Message-ID: First, I like new commit window/sheet, very nice. Second. it seems I cannot close the sheet with a keyboard shortcut if there's a commit message present. Is that intentional? Note that when amending there's always a commit message. -- /Jacob Carlborg From mailinglist at textmate.org Thu Sep 18 08:32:38 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Thu, 18 Sep 2014 10:32:38 +0200 Subject: [TxMt] Re: Closing the new commit window In-Reply-To: References: Message-ID: On 18 Sep 2014, at 10:10, Jacob Carlborg wrote: > [?] it seems I cannot close the sheet with a keyboard shortcut if > there's a commit message present. Is that intentional? I can cancel with ?. and commit with ??. Is that working for you? > Note that when amending there's always a commit message. Which is such an awesome feature, kudos to Ronald Wampler for implementing this (and all the commit window improvements). From mailinglist at textmate.org Thu Sep 18 08:37:48 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Thu, 18 Sep 2014 10:37:48 +0200 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> <7C268632-2F53-407F-A521-7A87A2EA4311@textmate.org> Message-ID: <87C02F38-D599-4A58-B6EF-EA055DE4DF5E@textmate.org> On 18 Sep 2014, at 10:04, Jacob Carlborg wrote: > [?] I also got the idea that TextMate is using PSMTabBarControl, > which MMTabBarView claims to be fairly compatible with. So I was > hoping it would be an easy change. Although I can be completely wrong. Ah, no. TextMate uses custom code which was supposed to be a re-usable control with declarative look and behavior. But it is a failed experiment (previously other things also used this control, but now only the tab bar view is left, and it?s more complex than it need to be, due to this attempt of generalization). > This suggestion might have come a bit late, I don't know for how long > you want to support the 10.9 look. I hope as long as 10.9 is > supported. Compared to iTerm which has the Yosemite look on 10.9, > which looks horrible. Using the 10.10 look on 10.9 would not make sense, so yeah, will retain the 10.9 look under 10.9. I had a brief look at MMTabBarView but I think its 10.6 compatibility requirement has made it not adopt a lot of new things that would make the implementation of this control simpler (and more flexible). I?ll spend some more time on this today and see where it takes us: Didn?t really intend to do tab bar view now (as there is actual broken stuff that needs fixing ;) ) but I?ve been wanting to get a bit more experience with layer-backed views? From samson.etienne at gmail.com Thu Sep 18 08:46:36 2014 From: samson.etienne at gmail.com (Etienne Samson) Date: Thu, 18 Sep 2014 10:46:36 +0200 Subject: [TxMt] Re: Bug in enclose selection with multi-key characters using US-International keyboard layout In-Reply-To: <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> References: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> Message-ID: <1A2FE56F-0511-4203-A4A0-8AE3DC34CFC7@gmail.com> Le 18 sept. 2014 ? 09:48, Allan Odgaard a ?crit : > On 17 Sep 2014, at 13:11, Eduardo Francos wrote: > >> When using the US-International keyboard layout, if you want to enclose selected text with quotes, single quotes or back-quotes by typing them, the originally selected text is deleted. > > I have both the ISO and ANSI US/international keyboard, and I can type all of the following quotes without using dead keys: ???, ???, "?", '?', and `?`. I can confirm that this also happens with the standard French layout with the ` backtick, the selected text just disappear. As the OP says, not a really harmful, significant bug, I just want to confirm that there's an issue. >> A really insignificant bug, I know :p and I can certainly live with it, so this is mostly for information. It's good to know someone knows it's there :-) > > I fear it would be rather complex to do a workaround for this because we would need to delay overtyping a selection when we detect the user starts a complex input sequence, and for most users I imagine this would be undesired (not seeing the selection immediately overtyped). A quick fix (IMHO, I'm no expert on the text system) would be to make dead keys be handled as normal keys as soon as you have something selected ? the rationale being that if you're actually trying to overtype a missed accent, you wouldn't rewrite a whole bunch of stuff. But there be dragons ;-). Regards, Etienne Samson -- samson.etienne at gmail.com From doob at me.com Thu Sep 18 09:03:34 2014 From: doob at me.com (Jacob Carlborg) Date: Thu, 18 Sep 2014 11:03:34 +0200 Subject: [TxMt] Re: Closing the new commit window In-Reply-To: References: Message-ID: On 18/09/14 10:32, Allan Odgaard wrote: > I can cancel with ?. and commit with ??. Is that working for you? Yes, that works. I was using escape before. Thanks. > Which is such an awesome feature, kudos to Ronald Wampler for > implementing this (and all the commit window improvements). Yeah, this is great, thanks. -- /Jacob Carlborg From mailinglist at textmate.org Thu Sep 18 09:13:19 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Thu, 18 Sep 2014 11:13:19 +0200 Subject: [TxMt] Re: Bug in enclose selection with multi-key characters using US-International keyboard layout In-Reply-To: <1A2FE56F-0511-4203-A4A0-8AE3DC34CFC7@gmail.com> References: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> <1A2FE56F-0511-4203-A4A0-8AE3DC34CFC7@gmail.com> Message-ID: <8ED8444D-A107-4419-A92F-76DFBC7443F6@textmate.org> On 18 Sep 2014, at 10:46, Etienne Samson wrote: >> I have both the ISO and ANSI US/international keyboard, and I can >> type all of the following quotes without using dead keys: ???, >> ???, "?", '?', and `?`. > > I can confirm that this also happens with the standard French layout > with the ` backtick, the selected text just disappear. As the OP says, > not a really harmful, significant bug, I just want to confirm that > there's an issue. Well, I can certainly reproduce it. What I didn?t understand is how the OP couldn?t type the above quotes on a US keyboard without using dead keys. > A quick fix (IMHO, I'm no expert on the text system) would be to make > dead keys be handled as normal keys as soon as you have something > selected [?] TextMate doesn?t know as much as you here. The input system is abstracted to support many different input modes. This is the protocol currently used to talk to TextMate: https://developer.apple.com/library/mac/documentation/cocoa/reference/applicationkit/Protocols/NSTextInput_Protocol/Reference/Reference.html As you can see, there is no ?here is a dead key that looks like a quote? but rather methods like -setMarkedText:selectedRange:. From mailinglist at textmate.org Thu Sep 18 09:16:46 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Thu, 18 Sep 2014 11:16:46 +0200 Subject: [TxMt] Re: Closing the new commit window In-Reply-To: References: Message-ID: On 18 Sep 2014, at 11:03, Jacob Carlborg wrote: >> I can cancel with ?. and commit with ??. Is that working for >> you? > > Yes, that works. I was using escape before. Thanks. I see ? escape only works when caret is not on a word, as it otherwise does buffer completion. So in the case of amending a commit, one would have to first use something like ?? (so getting used to ?. is probably better). From doob at me.com Thu Sep 18 09:23:51 2014 From: doob at me.com (Jacob Carlborg) Date: Thu, 18 Sep 2014 11:23:51 +0200 Subject: [TxMt] Re: Yosemite view effects In-Reply-To: <87C02F38-D599-4A58-B6EF-EA055DE4DF5E@textmate.org> References: <9C456B9A-1875-4AF1-A465-70A5C01219C9@igorkozlov.me> <5D041DA7-662F-4073-8EAE-FC89829B582A@icloud.com> <3092E1C6-0170-4C92-924B-CB7631F22DDC@textmate.org> <7C268632-2F53-407F-A521-7A87A2EA4311@textmate.org> <87C02F38-D599-4A58-B6EF-EA055DE4DF5E@textmate.org> Message-ID: On 18/09/14 10:37, Allan Odgaard wrote: > Ah, no. TextMate uses custom code which was supposed to be a re-usable > control with declarative look and behavior. But it is a failed > experiment (previously other things also used this control, but now only > the tab bar view is left, and it?s more complex than it need to be, due > to this attempt of generalization). I see. > Using the 10.10 look on 10.9 would not make sense, so yeah, will retain > the 10.9 look under 10.9. That's relieve. > I had a brief look at MMTabBarView but I think its 10.6 compatibility > requirement has made it not adopt a lot of new things that would make > the implementation of this control simpler (and more flexible). I see. I haven't use MMTabBarView myself. I just noticed it was a pre-existing tab control that supports the Safari look. > I?ll spend some more time on this today and see where it takes us: Didn?t > really intend to do tab bar view now (as there is actual broken stuff > that needs fixing ;) ) but I?ve been wanting to get a bit more > experience with layer-backed views? Cool, good luck :) -- /Jacob Carlborg From doob at me.com Thu Sep 18 09:30:26 2014 From: doob at me.com (Jacob Carlborg) Date: Thu, 18 Sep 2014 11:30:26 +0200 Subject: [TxMt] Re: Closing the new commit window In-Reply-To: References: Message-ID: On 18/09/14 11:16, Allan Odgaard wrote: > I see ? escape only works when caret is not on a word, as it otherwise > does buffer completion. Right, of course. I didn't think of that. I guess it worked previously when the window had a plain text field and not a TextMate editor view. > So in the case of amending a commit, one would > have to first use something like ?? (so getting used to ?. is probably > better). Yeah, I'll do that, now when I know about it :) -- /Jacob Carlborg From dipnlik at gmail.com Thu Sep 18 11:39:43 2014 From: dipnlik at gmail.com (dipnlik) Date: Thu, 18 Sep 2014 08:39:43 -0300 Subject: [TxMt] Re: Bug in enclose selection with multi-key characters using US-International keyboard layout In-Reply-To: <8ED8444D-A107-4419-A92F-76DFBC7443F6@textmate.org> References: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> <1A2FE56F-0511-4203-A4A0-8AE3DC34CFC7@gmail.com> <8ED8444D-A107-4419-A92F-76DFBC7443F6@textmate.org> Message-ID: On Thu, Sep 18, 2014 at 6:13 AM, Allan Odgaard wrote: > On 18 Sep 2014, at 10:46, Etienne Samson wrote: > >>> I have both the ISO and ANSI US/international keyboard, and I can type >>> all of the following quotes without using dead keys: ???, ???, "?", '?', and >>> `?`. >> >> >> I can confirm that this also happens with the standard French layout with >> the ` backtick, the selected text just disappear. As the OP says, not a >> really harmful, significant bug, I just want to confirm that there's an >> issue. > > > Well, I can certainly reproduce it. What I didn?t understand is how the OP > couldn?t type the above quotes on a US keyboard without using dead keys. I use the US-International keyboard layout too. To type ', I usually press ' then space, but I just accidentally discovered I can press ctrl ' for that. If I select text then try quoting it via ' then space, I trigger the reported bug, but not when using ctrl '. So, OP can consider using ctrl ' for quoting as a workaround. Also helpful: Toggle Single / Double String Quotes from the Source bundle, and Toggle Quote Style from the Ruby bundle (both accessible via ctrl cmd T). To type ", I press shift ' then space. I don't know how to type this character any other way to act as a workaround for quoting selected text. Also, I don't know how to type ?, ?, ? or ?, it would be nice to learn how to do that. -- :: dip -- From me at igorkozlov.me Thu Sep 18 11:54:04 2014 From: me at igorkozlov.me (Igor K) Date: Thu, 18 Sep 2014 14:54:04 +0300 Subject: [TxMt] Pressing "O" in Git Blame Message-ID: <44B4397C-0157-4C7D-9535-A779D12DABA5@igorkozlov.me> Textmate gets frozen for 15 seconds and then throws ``` JavaScript Warning The command ?ruby18 "/Users/igor1043k/Library/Application Support/TextMate/Managed/Bundles/Git.tmbundle/Support"/dispatch.rb "controller=log" "action=open_revision" "file_path=/Users/igor1043k/git/Bilbeo/Application/DataAPI/libs/models/Workspace.js" "revision=9201f304"? has been running for 15 seconds. Would you like to stop it? To avoid this warning, the bundle command should use the asynchronous version of TextMate.system(). ``` This bug is very old and it looks like no one but me uses this feature. Maybe someone with ruby knowledge can fix this thing? Thanks. From realoreocookie at gmx.de Thu Sep 18 12:04:38 2014 From: realoreocookie at gmx.de (Max Lein) Date: Thu, 18 Sep 2014 08:04:38 -0400 Subject: [TxMt] Search & Replace text field visibility bug back in alpha.9565 Message-ID: A while ago, I reported a text field visibility bug in the Search & Replace window, and after updating to alpha.9565, I noticed it?s back. When you bring up the Search & Replace window, anything you type in the search or replace text fields will not appear until you bring up the Search & Replace window again (e. g. by aborting and pressing Command + F again). Does anyone else see that behavior? Max From caiofbpa at icloud.com Thu Sep 18 12:54:28 2014 From: caiofbpa at icloud.com (Caio Fernando Bertoldi Paes de Andrade) Date: Thu, 18 Sep 2014 09:54:28 -0300 Subject: [TxMt] Re: Closing the new commit window In-Reply-To: References: Message-ID: <032C9BEE-6E38-4413-AECC-2A291DAE11D0@icloud.com> Jacob, I press Cmd+W to cancel the commit, it?s much more intuitive than Cmd+. to me, because Cmd+W closes tabs too. Caio > On 18 of Sep, 2014, at 06:30, Jacob Carlborg wrote: > > On 18/09/14 11:16, Allan Odgaard wrote: > >> I see ? escape only works when caret is not on a word, as it otherwise >> does buffer completion. > > Right, of course. I didn't think of that. I guess it worked previously when the window had a plain text field and not a TextMate editor view. > >> So in the case of amending a commit, one would >> have to first use something like ?? (so getting used to ?. is probably >> better). > > Yeah, I'll do that, now when I know about it :) > > -- > /Jacob Carlborg > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate From me at igorkozlov.me Thu Sep 18 12:58:47 2014 From: me at igorkozlov.me (Igor K) Date: Thu, 18 Sep 2014 15:58:47 +0300 Subject: [TxMt] Re: Closing the new commit window In-Reply-To: <032C9BEE-6E38-4413-AECC-2A291DAE11D0@icloud.com> References: <032C9BEE-6E38-4413-AECC-2A291DAE11D0@icloud.com> Message-ID: ?W doesn?t work for me for commit window > On 18 09 2014, at 15:54, Caio Fernando Bertoldi Paes de Andrade wrote: > > Jacob, > > I press Cmd+W to cancel the commit, it?s much more intuitive than Cmd+. to me, because Cmd+W closes tabs too. > > Caio > >> On 18 of Sep, 2014, at 06:30, Jacob Carlborg wrote: >> >> On 18/09/14 11:16, Allan Odgaard wrote: >> >>> I see ? escape only works when caret is not on a word, as it otherwise >>> does buffer completion. >> >> Right, of course. I didn't think of that. I guess it worked previously when the window had a plain text field and not a TextMate editor view. >> >>> So in the case of amending a commit, one would >>> have to first use something like ?? (so getting used to ?. is probably >>> better). >> >> Yeah, I'll do that, now when I know about it :) >> >> -- >> /Jacob Carlborg >> >> >> _______________________________________________ >> textmate mailing list >> textmate at lists.macromates.com >> http://lists.macromates.com/listinfo/textmate > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate From realoreocookie at gmx.de Thu Sep 18 14:01:19 2014 From: realoreocookie at gmx.de (Max Lein) Date: Thu, 18 Sep 2014 10:01:19 -0400 Subject: [TxMt] Re: Search & Replace text field visibility bug back in alpha.9565 Message-ID: After updating to 10.9.5, the issue seems to have been fixed. Max From rdwampler at gmail.com Thu Sep 18 20:06:32 2014 From: rdwampler at gmail.com (Ronald Wampler) Date: Thu, 18 Sep 2014 16:06:32 -0400 Subject: [TxMt] Re: Pressing "O" in Git Blame In-Reply-To: <44B4397C-0157-4C7D-9535-A779D12DABA5@igorkozlov.me> References: <44B4397C-0157-4C7D-9535-A779D12DABA5@igorkozlov.me> Message-ID: > On Thu, Sep 18, 2014 at 7:54 AM, Igor K wrote: > > > This bug is very old and it looks like no one but me uses this feature. > Maybe someone with ruby knowledge can fix this thing? > I submitted a PR (https://github.com/textmate/git.tmbundle/pull/31) that should resolve this. Can you test to confirm? -------------- next part -------------- An HTML attachment was scrubbed... URL: From me at igorkozlov.me Thu Sep 18 20:24:47 2014 From: me at igorkozlov.me (Igor K) Date: Thu, 18 Sep 2014 23:24:47 +0300 Subject: [TxMt] Re: Pressing "O" in Git Blame In-Reply-To: References: <44B4397C-0157-4C7D-9535-A779D12DABA5@igorkozlov.me> Message-ID: <1EFAB862-5835-4B95-A2D9-766F910E1D0F@igorkozlov.me> Works for me! Thanks! > On 18 09 2014, at 23:06, Ronald Wampler wrote: > > > On Thu, Sep 18, 2014 at 7:54 AM, Igor K > wrote: > > > This bug is very old and it looks like no one but me uses this feature. Maybe someone with ruby knowledge can fix this thing? > > I submitted a PR (https://github.com/textmate/git.tmbundle/pull/31 ) that should resolve this. Can you test to confirm? > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From efrancos at efrancos.com Fri Sep 19 14:16:00 2014 From: efrancos at efrancos.com (Eduardo Francos) Date: Fri, 19 Sep 2014 16:16:00 +0200 Subject: [TxMt] Re: Bug in enclose selection with multi-key characters using US-International keyboard layout In-Reply-To: <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> References: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> Message-ID: On 18 sept. 2014, at 09:48, Allan Odgaard wrote: > On 17 Sep 2014, at 13:11, Eduardo Francos wrote: > >> When using the US-International keyboard layout, if you want to enclose selected text with quotes, single quotes or back-quotes by typing them, the originally selected text is deleted. > > I have both the ISO and ANSI US/international keyboard, and I can type all of the following quotes without using dead keys: ???, ???, "?", '?', and `?`. My bad. I forgot I had installed the US International PC layout years ago using one that implements the Windows US International layout logic for diacritical characters. Apple's own US Extended layout does not have this problem. I may even switch to it, thanks for the tip. But that still leaves other layouts with diacritical accents. >> A really insignificant bug, I know :p and I can certainly live with it, so this is mostly for information. It's good to know someone knows it's there :-) > > I fear it would be rather complex to do a workaround for this because we would need to delay overtyping a selection when we detect the user starts a complex input sequence, and for most users I imagine this would be undesired (not seeing the selection immediately overtyped). > > I think it?d be fair to say that characters used for auto-pairing have the same limitations as characters available for key equivalents, where your issue also exist, i.e. you can?t bind a menu item to characters that you cannot type with a single key sans modifiers Seems reasonable. As I said the issue is insignificant and I, as probably others experiencing this problem, have long since learned ways to overcome it. Regards, Gualo -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 495 bytes Desc: Message signed with OpenPGP using GPGMail URL: From efrancos at efrancos.com Fri Sep 19 15:28:05 2014 From: efrancos at efrancos.com (Eduardo Francos) Date: Fri, 19 Sep 2014 17:28:05 +0200 Subject: [TxMt] Re: Bug in enclose selection with multi-key characters using US-International keyboard layout In-Reply-To: <8ED8444D-A107-4419-A92F-76DFBC7443F6@textmate.org> References: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> <1A2FE56F-0511-4203-A4A0-8AE3DC34CFC7@gmail.com> <8ED8444D-A107-4419-A92F-76DFBC7443F6@textmate.org> Message-ID: <45B9FED5-7012-4DB3-984A-A13C338035B4@efrancos.com> On 18 sept. 2014, at 11:13, Allan Odgaard wrote: > On 18 Sep 2014, at 10:46, Etienne Samson wrote: > ... > TextMate doesn?t know as much as you here. The input system is abstracted to support many different input modes. This is the protocol currently used to talk to TextMate: https://developer.apple.com/library/mac/documentation/cocoa/reference/applicationkit/Protocols/NSTextInput_Protocol/Reference/Reference.html > > As you can see, there is no ?here is a dead key that looks like a quote? but rather methods like -setMarkedText:selectedRange:. After some searching I found that actually there is. It's related to the 'setMarkedText', 'insertText', 'hasMarkedText' and 'unmarkText' selectors. Apparently the marked text is text that is pending input to be finished before the final modifications to the text are applied. Look at this short discussion in a Cocobuilder forum (http://www.cocoabuilder.com/archive/cocoa/144153-nstextinput-protocol-unmarktext-is-late.html) It seems from the program's behavior that TM's keyboard<->action binding system activates the action as soon as it receives the initial dead key character and before the sequence of characters is finished (hasMarkedText would/should return TRUE). If it could wait until all the insertText calls are issued followed by the unmarkText call then it would have the real character intended by the user. I dived in the code (mainly OakTextView) and couldn't understand what was going on. It seems to me that the marked text is being used for handling multiple selections or something. Anyway, hope this helps. -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 495 bytes Desc: Message signed with OpenPGP using GPGMail URL: From mailinglist at textmate.org Fri Sep 19 16:31:25 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Fri, 19 Sep 2014 18:31:25 +0200 Subject: [TxMt] Re: Bug in enclose selection with multi-key characters using US-International keyboard layout In-Reply-To: <45B9FED5-7012-4DB3-984A-A13C338035B4@efrancos.com> References: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> <1A2FE56F-0511-4203-A4A0-8AE3DC34CFC7@gmail.com> <8ED8444D-A107-4419-A92F-76DFBC7443F6@textmate.org> <45B9FED5-7012-4DB3-984A-A13C338035B4@efrancos.com> Message-ID: <7F4DD7E1-5588-4E3C-9256-963F15401A45@textmate.org> On 19 Sep 2014, at 17:28, Eduardo Francos wrote: > [?] It seems from the program's behavior that TM's keyboard<->action > binding system activates the action as soon as it receives the initial > dead key character and before the sequence of characters is finished > (hasMarkedText would/should return TRUE). If it could wait until all > the insertText calls are issued followed by the unmarkText call then > it would have the real character intended by the user. I don?t think this is the problem, rather, in the case of typing ? via a dead key, the user first types ? and the system asks TextMate to insert ? and have it *selected*, this is where TextMate will overwrite your current selection (and make a new), but since it?s part of a multi-input sequence, it won?t really do more. Next the user press E and the system then asks TextMate to insert ?, this time not as selected, but as we already have ? as selected text, it?ll replace what?s already typed (and the original selection you want to wrap was lost in first step). > I dived in the code (mainly OakTextView) and couldn't understand what > was going on [?] I can?t claim to fully understand Apple?s NSTextInput protocol and the current code is the result of lots of trial-and-error with various official and third party input modes, so I?m not keen on touching it to try and address your issue, not to say it?s impossible. The NSTextInput protocol has since been superseded, and at some time I will update the code to use the new protocol, I?ll keep your issue in mind when updating it. From toru.mori at coupa.com Fri Sep 19 21:08:00 2014 From: toru.mori at coupa.com (Toru Mori) Date: Fri, 19 Sep 2014 14:08:00 -0700 Subject: [TxMt] YAML bundle doesn't work for fold/unfold Message-ID: Sorry if this is FAQ, but can you please point me to where I should go? I'm using TextMate version 2.0-alpha.9561, but this issue persists for more than a year and Googling didn't give me any resolution yet. Simply, it doesn't fold/unfold for YAML file. In 'Edit Bundles', 'Settings' doesn't include 'Folding' while other bundles such as XML does. I guess foldingStopMarker might be wrong, but don't know if it's really used since it doesn't show up in 'Edit Budles / Settings'. Thanks in advance, Toru -------------- next part -------------- An HTML attachment was scrubbed... URL: From ryan at ryanfitzer.com Fri Sep 19 22:06:08 2014 From: ryan at ryanfitzer.com (Ryan Fitzer) Date: Fri, 19 Sep 2014 15:06:08 -0700 Subject: [TxMt] Questions on Search/Find Interaction Message-ID: 1. Is there a way to use the current selection for "Incremental Search" (?S)? It would be great to have "Use Selection for Find" (?E) to do the same for "Incremental Search". 2. When using the combo of "Use Selection for Find" (?E) and "Find All" (??F), the file is scrolled to the last match. If you're at the top of a long file, having it scrolled to the bottom can be pretty disorienting. Is there a way to reverse this to have the file scrolled to the first match? 3. Using the same combo as before, ?E then ??F, is there a shortcut for jumping through each match without deselecting the matches? 4. Is there a way to enable "Incremental Search" (?S) to highlight all matches as you type, instead of just the first? Thanks, Ryan -------------- next part -------------- An HTML attachment was scrubbed... URL: From mailinglist at textmate.org Sat Sep 20 08:17:19 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Sat, 20 Sep 2014 10:17:19 +0200 Subject: [TxMt] Re: Bug in enclose selection with multi-key characters using US-International keyboard layout In-Reply-To: <45B9FED5-7012-4DB3-984A-A13C338035B4@efrancos.com> References: <8460A339-99A6-4AD8-854F-B1DC2EC7F5A9@efrancos.com> <2450CB68-C874-4C0B-9717-4DC884971C06@textmate.org> <1A2FE56F-0511-4203-A4A0-8AE3DC34CFC7@gmail.com> <8ED8444D-A107-4419-A92F-76DFBC7443F6@textmate.org> <45B9FED5-7012-4DB3-984A-A13C338035B4@efrancos.com> Message-ID: <230B486D-A193-4705-81C8-CCD92CAC3FA2@textmate.org> On 19 Sep 2014, at 17:28, Eduardo Francos wrote: > After some searching [?] It occurred to me that you might get your desired behavior by binding a custom action to the key for the case where there is a selection. E.g. make a snippet with: ${0:?$TM_SELECTED_TEXT?} Set the key equivalent to (assuming US keyboard): ?E Set the scope selector to: dyn.selection After this, pressing ?E with a selection wraps the selection in acute accents (and keeps it selected). If there is no selection, it should insert ? (as ?marked?, allowing you to e.g. type E to make it ?). From mailinglist at textmate.org Sat Sep 20 08:26:18 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Sat, 20 Sep 2014 10:26:18 +0200 Subject: [TxMt] Re: YAML bundle doesn't work for fold/unfold In-Reply-To: References: Message-ID: <8D710527-3AD0-49B9-AEB0-E83197C2EA79@textmate.org> On 19 Sep 2014, at 23:08, Toru Mori wrote: > Simply, it doesn't fold/unfold for YAML file. In 'Edit Bundles', > 'Settings' > doesn't include 'Folding' while other bundles such as XML does. > I guess foldingStopMarker might be wrong, but don't know if it's > really > used since it doesn't show up in 'Edit Budles / Settings'. The TextMate 1.x folding system requires an explicit start and stop marker for foldings. The system is documented here: http://manual.macromates.com/en/navigation_overview#customizing_foldings I believe that YAML is based entirely on indent, so this system cannot work with YAML. In TextMate 2.0 an indented folding system was introduced, this is used for foldings in Python, Markdown headings, and C++ accessor keywords, not sure if it?s used elsewhere. It is mentioned in the ?Fold on the Dotted Line? section here http://blog.macromates.com/2012/the-layout-engine/ If you can come up with a pattern to use for foldingIndentedBlockStart then we?ll gladly add it to the YAML bundle. From mailinglist at textmate.org Sat Sep 20 08:38:23 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Sat, 20 Sep 2014 10:38:23 +0200 Subject: [TxMt] Re: Questions on Search/Find Interaction In-Reply-To: References: Message-ID: On 20 Sep 2014, at 0:06, Ryan Fitzer wrote: > 3. Using the same combo as before, ?E then ??F, is there a > shortcut for > jumping through each match without deselecting the matches? If you press ?W then it will select the current word, but if there already is a selection, then it?ll extend the selection to include the next occurrence of what?s currently selected. This is a very powerful feature that probably gets very little mention, not only to quickly replace multiple instances of the same thing, but also to quickly get multiple carets in locations that are not otherwise easy to get (via the other ways to break into multiple carets). Take this snippet: if(!something) { performActionY(); } else { performActionX(); } Say we want to get rid of the negation and swap the two blocks. We could use copy/paste with a lot of arrow up/down, since we first have to copy the block, then the ?else? line. Alternatively, select first open-brace, press ?W, and then press ?. Now we have a caret in each block: if(!something) {? performActionY(); } else {? performActionX(); } To swap the blocks, press ??B (Edit ? Select ? Enclosing Typing Pairs) and then swap with ?T (Text ? Transpose). The Transpose feature is another action that deserves more mention as it?s quite versatile, but I?ll leave that for another post. As for your other questions, I?m afraid the answer is no to those. I don?t use the incremental search myself, which means it?s generally last on the list of things I?d work on, but also that I?d be open to pull requests to change its behavior (since I don?t really care how it works). -------------- next part -------------- An HTML attachment was scrubbed... URL: From carpii.uk at gmail.com Sun Sep 21 21:59:03 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Sun, 21 Sep 2014 22:59:03 +0100 Subject: [TxMt] Keyboard based block-selection? Message-ID: If I hold down ALT, I can use the mouse to select a rectangular block of text Just wondering if there is any way to do this using solely the keyboard? Thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: From sellmerfud at gmail.com Sun Sep 21 22:05:43 2014 From: sellmerfud at gmail.com (Curt Sellmer) Date: Sun, 21 Sep 2014 17:05:43 -0500 Subject: [TxMt] Re: Keyboard based block-selection? In-Reply-To: References: Message-ID: Start a selection by any means (such as Shift-Right Arrow). Then tap the Option/Alt key. This will toggle between block selection and regular selection. Now you can hold the shift key down and use the arrow keys to extend the block selection. On Sun, Sep 21, 2014 at 4:59 PM, Carpii UK wrote: > If I hold down ALT, I can use the mouse to select a rectangular block of > text > > Just wondering if there is any way to do this using solely the keyboard? > > Thanks > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From carpii.uk at gmail.com Sun Sep 21 22:06:47 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Sun, 21 Sep 2014 23:06:47 +0100 Subject: [TxMt] Re: Keyboard based block-selection? In-Reply-To: References: Message-ID: excellent, thanks so much :) On 21 September 2014 23:05, Curt Sellmer wrote: > Start a selection by any means (such as Shift-Right Arrow). Then tap the > Option/Alt key. This will toggle between block selection and regular > selection. Now you can hold the shift key down and use the arrow keys to > extend the block selection. > > On Sun, Sep 21, 2014 at 4:59 PM, Carpii UK wrote: > >> If I hold down ALT, I can use the mouse to select a rectangular block of >> text >> >> Just wondering if there is any way to do this using solely the keyboard? >> >> Thanks >> >> >> _______________________________________________ >> textmate mailing list >> textmate at lists.macromates.com >> http://lists.macromates.com/listinfo/textmate >> > > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From ryan at ryanfitzer.com Sun Sep 21 23:20:18 2014 From: ryan at ryanfitzer.com (Ryan Fitzer) Date: Sun, 21 Sep 2014 16:20:18 -0700 Subject: [TxMt] Re: Questions on Search/Find Interaction In-Reply-To: References: Message-ID: Thanks for the in-depth example! Didn't realize ^W would select the next occurrence of the current selection. This is a great alternative for the combo ?E then ??F, when in short files. On Sat, Sep 20, 2014 at 1:38 AM, Allan Odgaard wrote: > On 20 Sep 2014, at 0:06, Ryan Fitzer wrote: > > > 1. Using the same combo as before, ?E then ??F, is there a shortcut > for jumping through each match without deselecting the matches? > > If you press ?W then it will select the current word, but if there > already is a selection, then it?ll extend the selection to include the next > occurrence of what?s currently selected. > > This is a very powerful feature that probably gets very little mention, > not only to quickly replace multiple instances of the same thing, but also > to quickly get multiple carets in locations that are not otherwise easy to > get (via the other ways to break into multiple carets). > > Take this snippet: > > if(!something) > { > performActionY(); > } > else > { > performActionX(); > } > > Say we want to get rid of the negation and swap the two blocks. We could > use copy/paste with a lot of arrow up/down, since we first have to copy the > block, then the ?else? line. > > Alternatively, select first open-brace, press ?W, and then press ?. Now we > have a caret in each block: > > if(!something) > {? > performActionY(); > } > else > {? > performActionX(); > } > > To swap the blocks, press ??B (Edit ? Select ? Enclosing Typing Pairs) and > then swap with ?T (Text ? Transpose). > > The Transpose feature is another action that deserves more mention as it?s > quite versatile, but I?ll leave that for another post. > > As for your other questions, I?m afraid the answer is no to those. I don?t > use the incremental search myself, which means it?s generally last on the > list of things I?d work on, but also that I?d be open to pull requests to > change its behavior (since I don?t really care how it works). > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From phil at bubblehouse.org Sun Sep 21 23:30:54 2014 From: phil at bubblehouse.org (Phil Christensen) Date: Sun, 21 Sep 2014 19:30:54 -0400 Subject: [TxMt] Re: Questions on Search/Find Interaction In-Reply-To: References: Message-ID: Alan/Ryan, Thanks much for this post. Not a very specific reason, except that now I finally get how to use Transpose. -phil > On Sep 20, 2014, at 4:38 AM, "Allan Odgaard" wrote: > > On 20 Sep 2014, at 0:06, Ryan Fitzer wrote: > > Using the same combo as before, ?E then ??F, is there a shortcut for jumping through each match without deselecting the matches? > If you press ?W then it will select the current word, but if there already is a selection, then it?ll extend the selection to include the next occurrence of what?s currently selected. > > This is a very powerful feature that probably gets very little mention, not only to quickly replace multiple instances of the same thing, but also to quickly get multiple carets in locations that are not otherwise easy to get (via the other ways to break into multiple carets). > > Take this snippet: > > if(!something) > { > performActionY(); > } > else > { > performActionX(); > } > Say we want to get rid of the negation and swap the two blocks. We could use copy/paste with a lot of arrow up/down, since we first have to copy the block, then the ?else? line. > > Alternatively, select first open-brace, press ?W, and then press ?. Now we have a caret in each block: > > if(!something) > {? > performActionY(); > } > else > {? > performActionX(); > } > To swap the blocks, press ??B (Edit ? Select ? Enclosing Typing Pairs) and then swap with ?T (Text ? Transpose). > > The Transpose feature is another action that deserves more mention as it?s quite versatile, but I?ll leave that for another post. > > As for your other questions, I?m afraid the answer is no to those. I don?t use the incremental search myself, which means it?s generally last on the list of things I?d work on, but also that I?d be open to pull requests to change its behavior (since I don?t really care how it works). > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From matt at tidbits.com Mon Sep 22 14:09:41 2014 From: matt at tidbits.com (Matt Neuburg) Date: Mon, 22 Sep 2014 07:09:41 -0700 Subject: [TxMt] jump back to previous location Message-ID: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> I have a feature suggestion. I've thought about this for years but it has only just occurred to me to say something about it. As you know, I use TextMate to write books. Big books with big chapters. So it often happens that I edit a chapter, then jump to an earlier / later bit of that same chapter to make some change entailed by what I was just writing. Now I want to "go back" to where I was before. That's the suggestion. TextMate should _automatically_ maintain "bookmarks" for the location of the two most recent edit locations (places where I actually typed or deleted text). Thus, I could use the bookmarks feature to jump back to where I was before. Just an idea. (A really great idea!) m. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From gerti-textmate at bitart.com Mon Sep 22 16:04:30 2014 From: gerti-textmate at bitart.com (Gerd Knops) Date: Mon, 22 Sep 2014 11:04:30 -0500 Subject: [TxMt] Re: jump back to previous location In-Reply-To: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: Seconded! A long time ago I used an editor that had this feature (Edison on Atari ST/TT), that was tremendously useful. I think I requested that feature several times over the years. If TM wouldn't be written mostly in C++ (gag) I long since would have added this and a whole slew of features. Gerd > On Sep 22, 2014, at 9:09 AM, Matt Neuburg wrote: > > I have a feature suggestion. I've thought about this for years but it has only just occurred to me to say something about it. > > As you know, I use TextMate to write books. Big books with big chapters. So it often happens that I edit a chapter, then jump to an earlier / later bit of that same chapter to make some change entailed by what I was just writing. Now I want to "go back" to where I was before. > > That's the suggestion. TextMate should _automatically_ maintain "bookmarks" for the location of the two most recent edit locations (places where I actually typed or deleted text). Thus, I could use the bookmarks feature to jump back to where I was before. > > Just an idea. (A really great idea!) m. > > -- > matt neuburg, phd = http://www.apeth.net/matt/ > pantes anthropoi tou eidenai oregontai phusei > Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do > iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do > RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html > > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate From elia at schito.me Mon Sep 22 16:07:10 2014 From: elia at schito.me (Elia Schito) Date: Mon, 22 Sep 2014 18:07:10 +0200 Subject: [TxMt] Re: jump back to previous location In-Reply-To: References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: +1 from here too I resort for the poor man's version (?Z / ??Z) all the time, but works only if there had been an edit of course Elia ? twitter/elia ?? github/elia On Mon, Sep 22, 2014 at 6:04 PM, Gerd Knops wrote: > Seconded! A long time ago I used an editor that had this feature (Edison > on Atari ST/TT), that was tremendously useful. > > I think I requested that feature several times over the years. If TM > wouldn't be written mostly in C++ (gag) I long since would have added this > and a whole slew of features. > > Gerd > > > On Sep 22, 2014, at 9:09 AM, Matt Neuburg wrote: > > > > I have a feature suggestion. I've thought about this for years but it > has only just occurred to me to say something about it. > > > > As you know, I use TextMate to write books. Big books with big chapters. > So it often happens that I edit a chapter, then jump to an earlier / later > bit of that same chapter to make some change entailed by what I was just > writing. Now I want to "go back" to where I was before. > > > > That's the suggestion. TextMate should _automatically_ maintain > "bookmarks" for the location of the two most recent edit locations (places > where I actually typed or deleted text). Thus, I could use the bookmarks > feature to jump back to where I was before. > > > > Just an idea. (A really great idea!) m. > > > > -- > > matt neuburg, phd = http://www.apeth.net/matt/ > > pantes anthropoi tou eidenai oregontai phusei > > Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do > > iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do > > RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html > > > > > > > > _______________________________________________ > > textmate mailing list > > textmate at lists.macromates.com > > http://lists.macromates.com/listinfo/textmate > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From timothy.c.bates at gmail.com Mon Sep 22 16:29:45 2014 From: timothy.c.bates at gmail.com (Timothy Bates) Date: Mon, 22 Sep 2014 09:29:45 -0700 (PDT) Subject: [TxMt] Re: jump back to previous location In-Reply-To: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: <1411403384936.06c42a95@Nodemailer> That is a GREAT idea, one which I sure we have all experienced at a latent level (?God I wish I could just go back to there now??) tricky to implement where ?there? is, but yes: let?s figure it out My current solution is Allan?s lovely new (ish?) show all finds.? But it?s an effort. What we need is something as intuitive as cntrl-W has turned out to be: you just ?go back to the last place you were editing stuff? with a simple ?return? keystroke? ^??-G? t ? Sent from Mailbox On Mon, Sep 22, 2014 at 3:09 PM, Matt Neuburg wrote: > I have a feature suggestion. I've thought about this for years but it has only just occurred to me to say something about it. > As you know, I use TextMate to write books. Big books with big chapters. So it often happens that I edit a chapter, then jump to an earlier / later bit of that same chapter to make some change entailed by what I was just writing. Now I want to "go back" to where I was before. > That's the suggestion. TextMate should _automatically_ maintain "bookmarks" for the location of the two most recent edit locations (places where I actually typed or deleted text). Thus, I could use the bookmarks feature to jump back to where I was before. > Just an idea. (A really great idea!) m. > -- > matt neuburg, phd = http://www.apeth.net/matt/ > pantes anthropoi tou eidenai oregontai phusei > Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do > iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do > RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From efrancos at efrancos.com Mon Sep 22 16:30:52 2014 From: efrancos at efrancos.com (Eduardo Francos) Date: Mon, 22 Sep 2014 18:30:52 +0200 Subject: [TxMt] Re: jump back to previous location In-Reply-To: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: <9C454651-1055-4171-B189-F6BDDE0F3258@efrancos.com> I developed a bundle I called "Navigator" for TM 1.5 that does that. Actually it does a lot more. Among others it implements a position stack with push/pop position, navigation within the stack, and coupled with its symbol tagging system it allows for an automatic call/return type of navigation to the symbol under the caret. If required it will open documents as needed. It also implements the function of project and global named bookmarks and navigation. I haven't had the time to update it to 2.0 but I should be working on it soon. Another problem to migrate it to TM 2.0 is that TMTOOLS is not supported anymore and I used it for precise positioning when documents have changed since the position was saved. Anyway, of you are patient I believe it could help you. On 22 sept. 2014, at 16:09, Matt Neuburg wrote: > > I have a feature suggestion. I've thought about this for years but it has only just occurred to me to say something about it. > > As you know, I use TextMate to write books. Big books with big chapters. So it often happens that I edit a chapter, then jump to an earlier / later bit of that same chapter to make some change entailed by what I was just writing. Now I want to "go back" to where I was before. > > That's the suggestion. TextMate should _automatically_ maintain "bookmarks" for the location of the two most recent edit locations (places where I actually typed or deleted text). Thus, I could use the bookmarks feature to jump back to where I was before. > > Just an idea. (A really great idea!) m. > > -- > matt neuburg, phd = http://www.apeth.net/matt/ > pantes anthropoi tou eidenai oregontai phusei > Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do > iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do > RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html > > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 495 bytes Desc: Message signed with OpenPGP using GPGMail URL: From mailinglist at textmate.org Mon Sep 22 16:41:47 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Mon, 22 Sep 2014 18:41:47 +0200 Subject: [TxMt] Re: jump back to previous location In-Reply-To: References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: I think the feature requested here is a bit fuzzy. Matt wants to do multiple edits in location A, then go to location B, do multiple edits, and have a way to get back to location A. Did I get that right? There is a current bookmark implementation, but you need to set them explicitly with ?F2 and can jump to next/previous with F2/?F2. Elia wants to ?scroll back? and says (about the poor man?s version) ?works only if there had been an edit?, so it seems where to scroll back to is ?where I was last, regardless of wether or not I did any edits?. So the definition of where to scroll back to is unclear to me. My interpretation of Gerd?s request (from earlier) is that each time the caret moves more than a line (after an edit), the position is added to the stack. But I am not sure if this would be a satisfying heuristic. There should probably be some collapsing of locations based on distance/within the view port. Gerd: If you do have a clear definition of how it should work, please let me know. On 22 Sep 2014, at 18:07, Elia Schito wrote: > +1 from here too > > I resort for the poor man's version (?Z / ??Z) all the time, but > works only > if there had been an edit of course > > Elia > > ? twitter/elia > ?? github/elia > > On Mon, Sep 22, 2014 at 6:04 PM, Gerd Knops > > wrote: > >> Seconded! A long time ago I used an editor that had this feature >> (Edison >> on Atari ST/TT), that was tremendously useful. >> >> I think I requested that feature several times over the years. If TM >> wouldn't be written mostly in C++ (gag) I long since would have added >> this >> and a whole slew of features. >> >> Gerd >> >>> On Sep 22, 2014, at 9:09 AM, Matt Neuburg wrote: >>> >>> I have a feature suggestion. I've thought about this for years but >>> it >> has only just occurred to me to say something about it. >>> >>> As you know, I use TextMate to write books. Big books with big >>> chapters. >> So it often happens that I edit a chapter, then jump to an earlier / >> later >> bit of that same chapter to make some change entailed by what I was >> just >> writing. Now I want to "go back" to where I was before. >>> >>> That's the suggestion. TextMate should _automatically_ maintain >> "bookmarks" for the location of the two most recent edit locations >> (places >> where I actually typed or deleted text). Thus, I could use the >> bookmarks >> feature to jump back to where I was before. >>> >>> Just an idea. (A really great idea!) m. >>> >>> -- >>> matt neuburg, phd = http://www.apeth.net/matt/ >>> pantes anthropoi tou eidenai oregontai phusei >>> Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do >>> iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do >>> RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html >>> >>> >>> >>> _______________________________________________ >>> textmate mailing list >>> textmate at lists.macromates.com >>> http://lists.macromates.com/listinfo/textmate >> >> >> _______________________________________________ >> textmate mailing list >> textmate at lists.macromates.com >> http://lists.macromates.com/listinfo/textmate >> > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate From mailinglist at textmate.org Mon Sep 22 16:50:02 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Mon, 22 Sep 2014 18:50:02 +0200 Subject: [TxMt] Re: jump back to previous location In-Reply-To: <9C454651-1055-4171-B189-F6BDDE0F3258@efrancos.com> References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> <9C454651-1055-4171-B189-F6BDDE0F3258@efrancos.com> Message-ID: On 22 Sep 2014, at 18:30, Eduardo Francos wrote: > I developed a bundle I called "Navigator" for TM 1.5 that does that. > Actually it does a lot more. Among others it implements a position > stack with push/pop position, navigation within the stack Sounds like this could be tackled by extended the current bookmark system with a ?go back to last bookmark added and remove it (and make it work accross files)?-action, and we would have this push/pop system. > [?] TMTOOLS is not supported anymore and I used it for precise > positioning when documents have changed since the position was saved. Curious, how was TMTOOLS able to help with that? From elia at schito.me Mon Sep 22 16:51:00 2014 From: elia at schito.me (Elia Schito) Date: Mon, 22 Sep 2014 18:51:00 +0200 Subject: [TxMt] Re: jump back to previous location In-Reply-To: References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> <9C454651-1055-4171-B189-F6BDDE0F3258@efrancos.com> Message-ID: My bad for shifting the target, sorry. I'll try to observe myself better and send a separate request that fits my usage (and possibly other's). Elia ? twitter/elia ?? github/elia On Mon, Sep 22, 2014 at 6:50 PM, Allan Odgaard wrote: > On 22 Sep 2014, at 18:30, Eduardo Francos wrote: > > I developed a bundle I called "Navigator" for TM 1.5 that does that. >> Actually it does a lot more. Among others it implements a position stack >> with push/pop position, navigation within the stack >> > > Sounds like this could be tackled by extended the current bookmark system > with a ?go back to last bookmark added and remove it (and make it work > accross files)?-action, and we would have this push/pop system. > > [?] TMTOOLS is not supported anymore and I used it for precise >> positioning when documents have changed since the position was saved. >> > > Curious, how was TMTOOLS able to help with that? > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From matt at tidbits.com Mon Sep 22 17:48:33 2014 From: matt at tidbits.com (Matt Neuburg) Date: Mon, 22 Sep 2014 10:48:33 -0700 Subject: [TxMt] Re: jump back to previous location In-Reply-To: References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: <5B20837B-0259-47DE-98DA-FF52D9F68B71@tidbits.com> On Sep 22, 2014, at 9:41 AM, Allan Odgaard wrote: > There is a current bookmark implementation, but you need to set them explicitly with ?F2 Right, of course. This is what I'm asking not to have to do. That's my point. Just _make_ me a bookmark wherever I am editing. Don't make this list too long; 2 levels would do (i.e. always keep autobookmarks at the 2 most recent places I edited). Example: on my GPS when I go running, there is an automatic waypoint set at my starting point, so if my run involves looping back, I can navigate to my starting place. Another example: In iBooks, if you tap a footnote marker to read the note, or use the chapter window to navigate to a different chapter, a "Back to p. xxx" button appears at the lower left. These are examples of how things in my world let me go elsewhere for a while and then return to where I was, automatically. TextMate should do that. m. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From gerti-textmate at bitart.com Mon Sep 22 17:52:49 2014 From: gerti-textmate at bitart.com (Gerd Knops) Date: Mon, 22 Sep 2014 12:52:49 -0500 Subject: [TxMt] Re: jump back to previous location In-Reply-To: References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: <1FFED256-146E-47F5-AB08-472AAA7B9CAC@bitart.com> > On Sep 22, 2014, at 11:41 AM, Allan Odgaard wrote: > > I think the feature requested here is a bit fuzzy. > [..] > > My interpretation of Gerd?s request (from earlier) is that each time the caret moves more than a line (after an edit), the position is added to the stack. But I am not sure if this would be a satisfying heuristic. There should probably be some collapsing of locations based on distance/within the view port. > > Gerd: If you do have a clear definition of how it should work, please let me know. > Certainly a fuzzy feature... Back then on small screens it was basically "moved more than a line". Now with large windows, it is trickier to find a good metric. Distance in the view port seems to factor in. Maybe this would work: still record every change in position that is more than a line, but then when going backward/forward in the history discard positions that are currently visible. Of course this may be too much, so it might need to be more like "visible and not more than y lines away from current cursor position". And I do think the "jump buffer" should span all files, not be limited to the one currently viewed. So say I work in file x, now go to look at something in file y (or even edit there) and "jump back", it should jump back to file x (even if I closed it!). And to add extra bells and whistles, there'd be an (optional) visual navigator that shows previous edit locations with context! Gerd > > On 22 Sep 2014, at 18:07, Elia Schito wrote: > >> +1 from here too >> >> I resort for the poor man's version (?Z / ??Z) all the time, but works only >> if there had been an edit of course >> >> Elia >> >> ? twitter/elia > >> ?? github/elia > >> >> On Mon, Sep 22, 2014 at 6:04 PM, Gerd Knops > >> wrote: >> >>> Seconded! A long time ago I used an editor that had this feature (Edison >>> on Atari ST/TT), that was tremendously useful. >>> >>> I think I requested that feature several times over the years. If TM >>> wouldn't be written mostly in C++ (gag) I long since would have added this >>> and a whole slew of features. >>> >>> Gerd >>> >>>> On Sep 22, 2014, at 9:09 AM, Matt Neuburg > wrote: >>>> >>>> I have a feature suggestion. I've thought about this for years but it >>> has only just occurred to me to say something about it. >>>> >>>> As you know, I use TextMate to write books. Big books with big chapters. >>> So it often happens that I edit a chapter, then jump to an earlier / later >>> bit of that same chapter to make some change entailed by what I was just >>> writing. Now I want to "go back" to where I was before. >>>> >>>> That's the suggestion. TextMate should _automatically_ maintain >>> "bookmarks" for the location of the two most recent edit locations (places >>> where I actually typed or deleted text). Thus, I could use the bookmarks >>> feature to jump back to where I was before. >>>> >>>> Just an idea. (A really great idea!) m. >>>> >>>> -- >>>> matt neuburg, phd = http://www.apeth.net/matt/ >>>> pantes anthropoi tou eidenai oregontai phusei >>>> Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do >>>> iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do >>>> RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html >>>> >>>> >>>> >>>> _______________________________________________ >>>> textmate mailing list >>>> textmate at lists.macromates.com >>>> http://lists.macromates.com/listinfo/textmate >>> >>> >>> _______________________________________________ >>> textmate mailing list >>> textmate at lists.macromates.com >>> http://lists.macromates.com/listinfo/textmate >>> >> >> _______________________________________________ >> textmate mailing list >> textmate at lists.macromates.com >> http://lists.macromates.com/listinfo/textmate > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From efrancos at efrancos.com Mon Sep 22 20:05:04 2014 From: efrancos at efrancos.com (Eduardo Francos) Date: Mon, 22 Sep 2014 22:05:04 +0200 Subject: [TxMt] Re: jump back to previous location In-Reply-To: References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> <9C454651-1055-4171-B189-F6BDDE0F3258@efrancos.com> Message-ID: <01C863B4-E126-49C1-80B6-586FF718302D@efrancos.com> On 22 sept. 2014, at 18:50, Allan Odgaard wrote: > On 22 Sep 2014, at 18:30, Eduardo Francos wrote: > >> I developed a bundle I called "Navigator" for TM 1.5 that does that. >> Actually it does a lot more. Among others it implements a position stack with push/pop position, navigation within the stack > > Sounds like this could be tackled by extended the current bookmark system with a ?go back to last bookmark added and remove it (and make it work accross files)?-action, and we would have this push/pop system. I agree. As long as you have a push/pop type of mechanism you have the functionality. However, current bookmarks are document based so if the document has been closed then the pop cannot work as is. Positioning bookmarks need to be kept available all the time. I use a file to keep the stack information between bundle calls. I guess this information could be managed more efficiently inside TM. The "push" could be done automatically to save the extra keystroke but as mentioned by others it's not easy to determine when a push must be done. For what it's worth, when I tried to implement automatic pushes I decided on two relatively simplistic conditions: 1) Whenever you edit a line its position is pushed. Previous pushed positions are poped and discarded while they are located within a number of lines of the current one. This in order to avoid the pain of going back to each one of them separately. 2) Whenever the caret stays on the same line or group of lines for longer than a given number of seconds the position is pushed. I translate this as "the position is interesting". > >> [?] TMTOOLS is not supported anymore and I used it for precise positioning when documents have changed since the position was saved. > > Curious, how was TMTOOLS able to help with that? It's related to the issue of files being modified after the position has been pushed so the stored position is not exactly right. After determining the file and position to jump to, and telling TM (via mate) to open the file at the position, I then use TMTOOLS to call a bundle command that would check that the line is still there. If it's not then search a (configurable) number of lines up and down to try to find it. This gives me precise positioning almost always. In a more general way I use TMTOOLS to be able to call back into TM and continue a logic that is dependent on TM doing something in between. It would be nice if the command line "mate" could do that type of things... but that's a different story :p > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- A non-text attachment was scrubbed... Name: signature.asc Type: application/pgp-signature Size: 495 bytes Desc: Message signed with OpenPGP using GPGMail URL: From carpii.uk at gmail.com Mon Sep 22 20:42:44 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Mon, 22 Sep 2014 21:42:44 +0100 Subject: [TxMt] Re: jump back to previous location In-Reply-To: <01C863B4-E126-49C1-80B6-586FF718302D@efrancos.com> References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> <9C454651-1055-4171-B189-F6BDDE0F3258@efrancos.com> <01C863B4-E126-49C1-80B6-586FF718302D@efrancos.com> Message-ID: I'd find this feature useful, even for coding. Personally I wouldn't want it to span files, or if it did, have the ability to disable that. I also wouldn't want it to re-purpose the existing bookmarks system, which I already use for navigating. But there's a lot of talk about how to detect line changes, and when to push a position onto the stack. It seems to me the easiest way would be just to backscan through the undo buffer, which can already tell us which lines have been changed on what line. Personally I would just keep it really simple, and have a key combo which backtraces through the undo buffer jumping to lines which were changed (without undoing anything obviously). As soon as you start adding timers, or refusing to jump to lines which are already visible, these things tend to just become unpredictable for the user, and end up being an annoyance rather than something useful. On 22 September 2014 21:05, Eduardo Francos wrote: > On 22 sept. 2014, at 18:50, Allan Odgaard > wrote: > > > On 22 Sep 2014, at 18:30, Eduardo Francos wrote: > > > >> I developed a bundle I called "Navigator" for TM 1.5 that does that. > >> Actually it does a lot more. Among others it implements a position > stack with push/pop position, navigation within the stack > > > > Sounds like this could be tackled by extended the current bookmark > system with a ?go back to last bookmark added and remove it (and make it > work accross files)?-action, and we would have this push/pop system. > > I agree. As long as you have a push/pop type of mechanism you have the > functionality. > However, current bookmarks are document based so if the document has been > closed then the pop cannot work as is. Positioning bookmarks need to be > kept available all the time. I use a file to keep the stack information > between bundle calls. > I guess this information could be managed more efficiently inside TM. > The "push" could be done automatically to save the extra keystroke but as > mentioned by others it's not easy to determine when a push must be done. > For what it's worth, when I tried to implement automatic pushes I decided > on two relatively simplistic conditions: > 1) Whenever you edit a line its position is pushed. Previous pushed > positions are poped and discarded while they are located within a number of > lines of the current one. This in order to avoid the pain of going back to > each one of them separately. > 2) Whenever the caret stays on the same line or group of lines for longer > than a given number of seconds the position is pushed. I translate this as > "the position is interesting". > > > > >> [?] TMTOOLS is not supported anymore and I used it for precise > positioning when documents have changed since the position was saved. > > > > Curious, how was TMTOOLS able to help with that? > > It's related to the issue of files being modified after the position has > been pushed so the stored position is not exactly right. After determining > the file and position to jump to, and telling TM (via mate) to open the > file at the position, I then use TMTOOLS to call a bundle command that > would check that the line is still there. If it's not then search a > (configurable) number of lines up and down to try to find it. This gives me > precise positioning almost always. > In a more general way I use TMTOOLS to be able to call back into TM and > continue a logic that is dependent on TM doing something in between. > It would be nice if the command line "mate" could do that type of > things... but that's a different story :p > > > > > > _______________________________________________ > > textmate mailing list > > textmate at lists.macromates.com > > http://lists.macromates.com/listinfo/textmate > > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From toru.mori at coupa.com Mon Sep 22 21:18:03 2014 From: toru.mori at coupa.com (Toru Mori) Date: Mon, 22 Sep 2014 14:18:03 -0700 Subject: [TxMt] Re: YAML bundle doesn't work for fold/unfold Message-ID: Thanks for the hint. I've experimented more than a couple of things and found the following settings works for me, at least it's much better than broken. Under 'Budles' => 'Edit' => 'Settings' => 'Indenting': { disableIndentCorrections = :true; indentOnPaste = 'simple'; foldingStartMarker = '^\s*"""(?=.)(?!.*""")'; foldingIndentedBlockStart = '^\s*.*(:|-) ?(&\w+)?(\{[^}"'']*|\([^)"'']*)?$'; foldingIndentedBlockIgnore = '^\s*#'; } foldingStartMarker was copied from Python setting. foldingIndentedBlockIgnore was copied from the 'Fold on the Dotted Line' section in http://blog.macromates.com/2012/the-layout-engine/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From matt at tidbits.com Mon Sep 22 22:48:06 2014 From: matt at tidbits.com (Matt Neuburg) Date: Mon, 22 Sep 2014 15:48:06 -0700 Subject: [TxMt] Re: jump back to previous location In-Reply-To: References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> <9C454651-1055-4171-B189-F6BDDE0F3258@efrancos.com> <01C863B4-E126-49C1-80B6-586FF718302D@efrancos.com> Message-ID: <50D3F5F9-32F5-4210-AC60-F030FECCCB0B@tidbits.com> On Sep 22, 2014, at 1:42 PM, Carpii UK wrote: > I also wouldn't want it to re-purpose the existing bookmarks system, which I already use for navigating. > I only suggested that because I thought it would make it easier (leverage existing interface / tech). m. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From erikn at myphysicslab.com Mon Sep 22 23:19:13 2014 From: erikn at myphysicslab.com (Erik Neumann) Date: Mon, 22 Sep 2014 16:19:13 -0700 Subject: [TxMt] Re: jump back to previous location In-Reply-To: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: I want this feature ALL THE TIME. There was some editor I used in the distant past that had this and it was super convenient. For me going between files would be helpful, not only within one file. Kind of like the back/forward buttons in a web browser. I appreciate it's probably complex to figure out how to do it, and satisfying everyone's particular flavor is impossible. Sorry I don't have time to give a more concrete suggestion at the moment. --ErikN On Mon, Sep 22, 2014 at 7:09 AM, Matt Neuburg wrote: > I have a feature suggestion. I've thought about this for years but it has > only just occurred to me to say something about it. > > As you know, I use TextMate to write books. Big books with big chapters. > So it often happens that I edit a chapter, then jump to an earlier / later > bit of that same chapter to make some change entailed by what I was just > writing. Now I want to "go back" to where I was before. > > That's the suggestion. TextMate should _automatically_ maintain > "bookmarks" for the location of the two most recent edit locations (places > where I actually typed or deleted text). Thus, I could use the bookmarks > feature to jump back to where I was before. > > Just an idea. (A really great idea!) m. > > -- > matt neuburg, phd = http://www.apeth.net/matt/ > pantes anthropoi tou eidenai oregontai phusei > Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do > iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do > RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html > > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From sanssecours at f-m.fm Tue Sep 23 09:19:45 2014 From: sanssecours at f-m.fm (sanssecours) Date: Tue, 23 Sep 2014 02:19:45 -0700 (PDT) Subject: [TxMt] Re: YAML bundle doesn't work for fold/unfold In-Reply-To: <8D710527-3AD0-49B9-AEB0-E83197C2EA79@textmate.org> References: <8D710527-3AD0-49B9-AEB0-E83197C2EA79@textmate.org> Message-ID: <1411463985655-28270.post@n5.nabble.com> Allan Odgaard-4 wrote > On 19 Sep 2014, at 23:08, Toru Mori wrote: > >> Simply, it doesn't fold/unfold for YAML file. In 'Edit Bundles', >> 'Settings' >> doesn't include 'Folding' while other bundles such as XML does. >> I guess foldingStopMarker might be wrong, but don't know if it's >> really >> used since it doesn't show up in 'Edit Budles / Settings'. > > The TextMate 1.x folding system requires an explicit start and stop > marker for foldings. The system is documented here: > http://manual.macromates.com/en/navigation_overview#customizing_foldings > > I believe that YAML is based entirely on indent, so this system cannot > work with YAML. > > In TextMate 2.0 an indented folding system was introduced, this is used > for foldings in Python, Markdown headings, and C++ accessor keywords, > not sure if it?s used elsewhere. > > It is mentioned in the ?Fold on the Dotted Line? section here > http://blog.macromates.com/2012/the-layout-engine/ > > If you can come up with a pattern to use for foldingIndentedBlockStart > then we?ll gladly add it to the YAML bundle. > > _______________________________________________ > textmate mailing list > textmate at .macromates > http://lists.macromates.com/listinfo/textmate Hi Allan and Toru, could it be that the current system for matching the end marker of foldings has problems with matching ?empty? lines? The folding marker used in the YAML bundle seems to work fine in TextMate 1 but not in TextMate 2. I attached two screenshots showing the different behaviour. Implementing the TextMate 1 behaviour inside TextMate 2 might fix the problem described by Toru. Kind regards, Ren? -- View this message in context: http://textmate.1073791.n5.nabble.com/YAML-bundle-doesn-t-work-for-fold-unfold-tp28245p28270.html Sent from the textmate users mailing list archive at Nabble.com. From bibiko at eva.mpg.de Tue Sep 23 12:15:34 2014 From: bibiko at eva.mpg.de (=?windows-1252?Q?Hans-J=F6rg_Bibiko?=) Date: Tue, 23 Sep 2014 14:15:34 +0200 Subject: [TxMt] Re: jump back to previous location In-Reply-To: References: <3E34B7D5-831B-4334-9E3A-2B3B0B1C29C0@tidbits.com> Message-ID: <86C54AA1-CDAA-462A-9C9F-4B56E2DDB2BB@eva.mpg.de> Hi, I ONLY would like to think loudly and !!quickly!!, thus please forgive me if I miss something essential If we?re talking ?only? about to jump to the last caret position (I?d argue to listen to a selection change rather than to an actual edit, because it often occurs that one goes to somewhere else to copy a chunk of text or read something only, ergo no actual edit happens) then this could be at least a possible starting point: Each document should store two records [last_MARKER, current_MARKER] to hold these 3 variables: - file_path (of the current doc) - current_selected_range [curRange] - range_of_visible_line_numbers [visibleRange] (first visible line number in view port (window) and length to the last visible line number) I define the circumstance whether a caret position has to be stored in a global stack (of size 2) or not via the first visible line number and the last visible line number of the current window plus the actual line number of the caret. In other words, each caret position (line number) which is not inside of the last remembered marker's first and last visible line numbers should be a candidate for the global stack. For a selection, do set [last_MARKER] only if both line numbers (begin and end of the selection) are inside of the last remembered marker?s visibleRange. Pseudo-Code: + init [last_MARKER] with defaults {for each doc} * listen to selectionDidChange notification + get line number of the begin of the current selection [curBegLineNr] + get line number of the end of the current selection [curEndLineNr] + set [current_MARKER] to current values if [curBegLineNr] inRange [last_MARKER.visibleRange] then // save the current status only if // the line number of the end of a possible selection // is also within the last range of visible line numbers // to avoid storing if user ?only? selected a // a large chunk - a next notification should handle this // or // if a doc change just happened if ([curEndLineNr] inRange [last_MARKER.visibleRange]) or (STACK.size > 0 AND STACK[0].file_path <> current_MARKER.file_path) then set [last_MARKER] to [current_MARKER] endif else // caret is outside of the last visible view port, // ergo add MARKER to global STACK set [last_MARKER] to [current_MARKER] move STACK[0] to STACK[1] and insert [last_MARKER] to STACK[0] endif * listen to active document will be changed + move STACK[0] to STACK[1] and insert [last_MARKER] of the not yet inactive doc to STACK[0] = navigation - Jump to last caret position A ?Jump to last caret position? command should take the last inserted MARKER.file_path and MARKER.curRange to restore that doc plus caret position AND swap both entries in the STACK in order to be able to jump to and fro between the both last positions. Remark: A stack size of 2 would have the advantage that one hasn?t to care about deletions, validations of ranges, etc. within docs, otherwise each update of the stack should check if all MARKERs are still valid. Cheers, Hans From textmate at floehopper.org Wed Sep 24 09:54:02 2014 From: textmate at floehopper.org (James Mead) Date: Wed, 24 Sep 2014 10:54:02 +0100 Subject: [TxMt] Viewport position lost when switching tabs with larger font sizes Message-ID: Steps to reproduce: * Set font size to larger than the default. * Position the cursor towards the bottom of a file such that the viewport is displaying the bottom part of that file. * Switch to another tab. * Switch back to the original tab. * See that although the cursor is still in the same place, the viewport is now displaying an earlier part of the file. It looks as if the viewport position calculation is not taking into account the larger font size. Cheers, James. From mailinglist at textmate.org Wed Sep 24 10:01:44 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Wed, 24 Sep 2014 12:01:44 +0200 Subject: [TxMt] Re: Viewport position lost when switching tabs with larger font sizes In-Reply-To: References: Message-ID: On 24 Sep 2014, at 11:54, James Mead wrote: > [?] It looks as if the viewport position calculation is not taking > into > account the larger font size. The problem is with how TextMate does lazy layout, it does an initial heuristic (fast) layout and uses that for scroll positions and then only does the real (slow) layout when needing to display the text, which can ?push things around? (and cause the previous correct scroll position to now be incorrect). Long-term I hope to improve this, but it?s non-trivial because laying out text is surprisingly slow. From tricon at me.com Wed Sep 24 23:36:16 2014 From: tricon at me.com (David Aaron Fendley) Date: Wed, 24 Sep 2014 16:36:16 -0700 Subject: [TxMt] Indenting Code & Markup Within Source Tags Message-ID: I would like code and markup inside certain tags to be indented. Currently, this occurs? Desired result: I?ve determined that this is because style, script, and other tags are marked as `source.*`. If I rename these language constructs to something that doesn?t start with ?source?, then indentation works as expected. However, I do not want to throwaway the advantages of inheriting from source. Does anyone know where I can begin looking to modify the `source` grammar so that code within certain tags is indented? I have not had any luck with `increaseIndentPattern` (yet). From tricon at me.com Wed Sep 24 23:37:22 2014 From: tricon at me.com (David Aaron Fendley) Date: Wed, 24 Sep 2014 16:37:22 -0700 Subject: [TxMt] Re: Indenting Code & Markup Within Source Tags In-Reply-To: References: Message-ID: Formatting did not transfer. In the second example, I?d like the `body` tag indented by two spaces. On Sep 24, 2014, at 4:36 PM, David Aaron Fendley wrote: > I would like code and markup inside certain tags to be indented. Currently, this occurs? > > > > Desired result: > > > > > I?ve determined that this is because style, script, and other tags are marked as `source.*`. If I rename these language constructs to something that doesn?t start with ?source?, then indentation works as expected. However, I do not want to throwaway the advantages of inheriting from source. > > Does anyone know where I can begin looking to modify the `source` grammar so that code within certain tags is indented? I have not had any luck with `increaseIndentPattern` (yet). > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate From robert.gomez.milton at gmail.com Thu Sep 25 02:08:48 2014 From: robert.gomez.milton at gmail.com (Robert Milton) Date: Thu, 25 Sep 2014 03:08:48 +0100 Subject: [TxMt] Re: Questions on Search/Find Interaction In-Reply-To: References: Message-ID: I?ll second that. This thread seems especially interesting and useful in a generic way. More like this please. On 22 Sep 2014, at 00:30, Phil Christensen wrote: > Alan/Ryan, > > Thanks much for this post. Not a very specific reason, except that now I finally get how to use Transpose. > > -phil > > On Sep 20, 2014, at 4:38 AM, "Allan Odgaard" wrote: > >> On 20 Sep 2014, at 0:06, Ryan Fitzer wrote: >> >> Using the same combo as before, ?E then ??F, is there a shortcut for jumping through each match without deselecting the matches? >> If you press ?W then it will select the current word, but if there already is a selection, then it?ll extend the selection to include the next occurrence of what?s currently selected. >> >> This is a very powerful feature that probably gets very little mention, not only to quickly replace multiple instances of the same thing, but also to quickly get multiple carets in locations that are not otherwise easy to get (via the other ways to break into multiple carets). >> >> Take this snippet: >> >> if(!something) >> { >> performActionY(); >> } >> else >> { >> performActionX(); >> } >> Say we want to get rid of the negation and swap the two blocks. We could use copy/paste with a lot of arrow up/down, since we first have to copy the block, then the ?else? line. >> >> Alternatively, select first open-brace, press ?W, and then press ?. Now we have a caret in each block: >> >> if(!something) >> {? >> performActionY(); >> } >> else >> {? >> performActionX(); >> } >> To swap the blocks, press ??B (Edit ? Select ? Enclosing Typing Pairs) and then swap with ?T (Text ? Transpose). >> >> The Transpose feature is another action that deserves more mention as it?s quite versatile, but I?ll leave that for another post. >> >> As for your other questions, I?m afraid the answer is no to those. I don?t use the incremental search myself, which means it?s generally last on the list of things I?d work on, but also that I?d be open to pull requests to change its behavior (since I don?t really care how it works). >> >> >> _______________________________________________ >> textmate mailing list >> textmate at lists.macromates.com >> http://lists.macromates.com/listinfo/textmate > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From svsdiscussion at gmail.com Thu Sep 25 16:43:49 2014 From: svsdiscussion at gmail.com (svsdiscussion at gmail.com) Date: Thu, 25 Sep 2014 09:43:49 -0700 (PDT) Subject: [TxMt] Soft Wrap in Commented Lines Message-ID: <1411663428998-28277.post@n5.nabble.com> When I enable soft wrap I get: On the other hand, when I disable soft wrap I get: I think the behaviour in disabled soft wrap when the line is commented is not correct. What do you think? -- View this message in context: http://textmate.1073791.n5.nabble.com/Soft-Wrap-in-Commented-Lines-tp28277.html Sent from the textmate users mailing list archive at Nabble.com. From doob at me.com Thu Sep 25 20:11:22 2014 From: doob at me.com (Jacob Carlborg) Date: Thu, 25 Sep 2014 22:11:22 +0200 Subject: [TxMt] Re: Soft Wrap in Commented Lines In-Reply-To: <1411663428998-28277.post@n5.nabble.com> References: <1411663428998-28277.post@n5.nabble.com> Message-ID: On 2014-09-25 18:43, svsdiscussion at gmail.com wrote: > When I enable soft wrap I get: > > On the other hand, when I disable soft wrap I get: > > I think the behaviour in disabled soft wrap when the line is commented is > not correct. > What do you think? It's a feature, see this [1] for how to disable it. [1] http://textmate.1073791.n5.nabble.com/Comments-ignore-quot-Soft-wrap-quot-settings-td27717.html -- /Jacob Carlborg From sanssecours at f-m.fm Thu Sep 25 20:21:53 2014 From: sanssecours at f-m.fm (sanssecours) Date: Thu, 25 Sep 2014 13:21:53 -0700 (PDT) Subject: [TxMt] Re: Soft Wrap in Commented Lines In-Reply-To: <1411663428998-28277.post@n5.nabble.com> References: <1411663428998-28277.post@n5.nabble.com> Message-ID: <1411676513690-28279.post@n5.nabble.com> Hi S.V., the behaviour you describe is actually intended. The Source bundle contains settings which always activate soft wrap for comments. If you disable ?Style: Line Comments - Misc.? you should get the behaviour you expected. For more information on the topic you can take a look at the F.A.Q. . Kind regards, Ren? -- View this message in context: http://textmate.1073791.n5.nabble.com/Soft-Wrap-in-Commented-Lines-tp28277p28279.html Sent from the textmate users mailing list archive at Nabble.com. From textmate at alxp.gr Fri Sep 26 13:57:53 2014 From: textmate at alxp.gr (enas) Date: Fri, 26 Sep 2014 15:57:53 +0200 Subject: [TxMt] crash on Find Next Message-ID: Hi, Here's a crash I'm getting deterministically on hitting ?G, when the text can't be found (when no occurrence exists, or after all occurrences have been found, even with the wrap around option selected). I can reproduce the crash as follows: - Create a new empty file, or open any existing text. - Open the Find dialog, enter a Find string that doesn't exist in the document, close the Find dialog. - Hit ?G. TextMate will always crash. I'm more than happy to provide any additional information, and my apologies if this has already been reported. enas -------------- next part -------------- An HTML attachment was scrubbed... URL: From mailinglist at textmate.org Fri Sep 26 14:36:01 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Fri, 26 Sep 2014 16:36:01 +0200 Subject: [TxMt] Re: crash on Find Next In-Reply-To: References: Message-ID: On 26 Sep 2014, at 15:57, enas wrote: > Here's a crash I'm getting deterministically on hitting ?G [?] Sorry about this, I had started to filter crash reports so I only looked at crashes from 10.9 (because there is so much noise from 10.10) and therefor missed that latest nightly broke tool tips for 10.7 users. I?ll push a fix later today. From bibiko at eva.mpg.de Sat Sep 27 16:41:36 2014 From: bibiko at eva.mpg.de (=?windows-1252?Q?Hans-J=F6rg_Bibiko?=) Date: Sat, 27 Sep 2014 18:41:36 +0200 Subject: [TxMt] [TM2] DIALOG2 replace input via HTML Javascript TextMate.system() Message-ID: <6BB1C760-BD74-44A9-A614-C67A7309FB75@eva.mpg.de> Hi, short question: Is it possible for a command (Input: ?Selection? - fallback to ?Line? ; Output: HTML window) to replace the "Input selection" via Javascript's TextMate.system()? TextMate.system("$DIALOG x-insert --text 'foo'", null) works but only for a selection and not for the fallback situation. Has anyone an idea? Thanks, Hans From mailinglist at textmate.org Sun Sep 28 08:44:45 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Sun, 28 Sep 2014 10:44:45 +0200 Subject: [TxMt] Re: [TM2] DIALOG2 replace input via HTML Javascript TextMate.system() In-Reply-To: <6BB1C760-BD74-44A9-A614-C67A7309FB75@eva.mpg.de> References: <6BB1C760-BD74-44A9-A614-C67A7309FB75@eva.mpg.de> Message-ID: On 27 Sep 2014, at 18:41, Hans-J?rg Bibiko wrote: > Is it possible for a command (Input: ?Selection? - fallback to > ?Line? ; Output: HTML window) to replace the "Input selection" via > Javascript's TextMate.system()? > > TextMate.system("$DIALOG x-insert --text 'foo'", null) works but only > for a selection and not for the fallback situation. > > Has anyone an idea? Untested, but you could run this if TM_SELECTED_TEXT is unset: TextMate.system('"$TM_MATE" -l"$TM_LINE_NUMBER-$((TM_LINE_NUMBER+1))"', null) Then it will select the current line and x-insert should thus replace it. By default "$TM_MATE" works on current document found via the TM_DOCUMENT_UUID environment variable. From bibiko at eva.mpg.de Sun Sep 28 09:19:06 2014 From: bibiko at eva.mpg.de (=?windows-1252?Q?Hans-J=F6rg_Bibiko?=) Date: Sun, 28 Sep 2014 11:19:06 +0200 Subject: [TxMt] Re: [TM2] DIALOG2 replace input via HTML Javascript TextMate.system() In-Reply-To: References: <6BB1C760-BD74-44A9-A614-C67A7309FB75@eva.mpg.de> Message-ID: <2769AC03-42B6-494C-8AF3-485E4BEBECAE@eva.mpg.de> On 28 Sep 2014, at 10:44, Allan Odgaard wrote: > On 27 Sep 2014, at 18:41, Hans-J?rg Bibiko wrote: > >> Is it possible for a command (Input: ?Selection? - fallback to ?Line? ; Output: HTML window) to replace the "Input selection" via Javascript's TextMate.system()? >> >> TextMate.system("$DIALOG x-insert --text 'foo'", null) works but only for a selection and not for the fallback situation. >> >> Has anyone an idea? > > Untested, but you could run this if TM_SELECTED_TEXT is unset: > > TextMate.system('"$TM_MATE" -l"$TM_LINE_NUMBER-$((TM_LINE_NUMBER+1))"', null) Haha - holy ..., it works like a charm :) Allan, thanks a lot, you saved my day! I've already played with "mate" but I didn't come up with the obvious one. If this works then I thought that this should also work: TextMate.system('"$TM_MATE" -l"$TM_LINE_NUMBER:1-$TM_LINE_NUMBER:1000000"', null) and yes :) - due to that this scripting language I'm working on does not allow long lines. So my line of code is now: TextMate.system('[[ -z $TM_SELECTED_TEXT ]] && "$TM_MATE" -l"$TM_LINE_NUMBER:1-$TM_LINE_NUMBER:1000000"', null); Kind regards, Hans From bibiko at eva.mpg.de Sun Sep 28 11:03:55 2014 From: bibiko at eva.mpg.de (=?windows-1252?Q?Hans-J=F6rg_Bibiko?=) Date: Sun, 28 Sep 2014 13:03:55 +0200 Subject: [TxMt] Re: [TM2] DIALOG2 replace input via HTML Javascript TextMate.system() In-Reply-To: <2769AC03-42B6-494C-8AF3-485E4BEBECAE@eva.mpg.de> References: <6BB1C760-BD74-44A9-A614-C67A7309FB75@eva.mpg.de> <2769AC03-42B6-494C-8AF3-485E4BEBECAE@eva.mpg.de> Message-ID: <062109B1-141F-4814-BA1D-3D74DD80A4F7@eva.mpg.de> On 28 Sep 2014, at 11:19, Hans-J?rg Bibiko wrote: > TextMate.system('[[ -z $TM_SELECTED_TEXT ]] && "$TM_MATE" -l"$TM_LINE_NUMBER:1-$TM_LINE_NUMBER:1000000"', null); Hmm, but this leads to another issue: How to get the actual _current_ line number since TM_QUERY or TM_LINE_NUMBER rely on the bash starting environment? And while the HTML output window is open the user can change the caret. Yet another idea? Best, Hans From carpii.uk at gmail.com Mon Sep 29 12:26:13 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Mon, 29 Sep 2014 13:26:13 +0100 Subject: [TxMt] Convert CRLF to LF Message-ID: I'm working with a legacy codebase, some of which uses CRLF line endings, mixed with more recent updates which use LF. Textmate displays these as Is there an existing bundle item which will strip the and make the full file LF consistent? I'd rather this was a manual action rather than Save As... or done transparently, because it can complicate git conflicts if its not managed. Thanks -------------- next part -------------- An HTML attachment was scrubbed... URL: From matt at tidbits.com Mon Sep 29 15:37:49 2014 From: matt at tidbits.com (Matt Neuburg) Date: Mon, 29 Sep 2014 08:37:49 -0700 Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: References: Message-ID: On Sep 29, 2014, at 5:26 AM, Carpii UK wrote: > Is there an existing bundle item which will strip the and make the full file LF consistent? > I use BBEdit (or its free little brother TextWrangler) for this. With BBEdit you can just create a text engine that batch-processes your files. m. -- matt neuburg, phd = http://www.apeth.net/matt/ pantes anthropoi tou eidenai oregontai phusei Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html From carpii.uk at gmail.com Mon Sep 29 16:05:27 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Mon, 29 Sep 2014 17:05:27 +0100 Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: References: Message-ID: Thanks, I was a paid up BBEdit user before I moved to TM2, but it fell out of favour after a few bugs (one of them was due to mismatched line endings funnily enough, it would often misreport line numbers) Im able to do it in TM by using Text -> Filter Through Command But a specific bundle item would be nice so I could assign a key combo and bypass the dialog (which often has other commands Ive used in it) On 29 September 2014 16:37, Matt Neuburg wrote: > > On Sep 29, 2014, at 5:26 AM, Carpii UK wrote: > > > Is there an existing bundle item which will strip the and make the > full file LF consistent? > > > > I use BBEdit (or its free little brother TextWrangler) for this. With > BBEdit you can just create a text engine that batch-processes your files. m. > > -- > matt neuburg, phd = http://www.apeth.net/matt/ > pantes anthropoi tou eidenai oregontai phusei > Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do > iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do > RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html > > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From bibiko at eva.mpg.de Mon Sep 29 16:07:34 2014 From: bibiko at eva.mpg.de (=?iso-8859-1?Q?Hans-J=F6rg_Bibiko?=) Date: Mon, 29 Sep 2014 18:07:34 +0200 Subject: [TxMt] Discussion: TextMate.save_if_untitled and TextMate::Executor.run -> version Message-ID: <6AA12EF7-6B07-41A2-BBF8-04817EBE2E92@eva.mpg.de> Hi, I'm developing a new bundle for a scripting language. For the "Run" command I'd like to make usage of TM's TextMate::Executor library. Fine. But I was confronted with two subjects for a discussion: 1) TextMate.save_if_untitled In order to use TextMate::Executor one should execute "TextMate.save_if_untitled('foo')" in beforehand. But I'd like to allow to the user that s/he runs either the entire document or only a selection. With the current implementation this doesn't work, so I suggest the following change (adding of the first 6 lines of code and a new argument): save_current_document.rb line: 33ff def save_if_untitled(temp_ext='tmp', ignore_selection=true) if not ignore_selection and not ENV['TM_SELECTED_TEXT'].nil? ENV['TM_FILEPATH'] = nil if not ENV['TM_SELECTED_TEXT'].nil? ENV['TM_DISPLAYNAME'] += ' (Selection)' end end return unless ENV['TM_FILEPATH'].nil? ENV['TM_FILEPATH'] = TextMate::IO.tempfile(temp_ext).path ENV['TM_FILENAME'] = File.basename(ENV['TM_FILEPATH']) ENV['TM_FILE_IS_UNTITLED'] = 'true' begin ...... This approach does't break the default behaviour. Only if the developer would like to add that functionality s/he has only set the second argument to "false". By adding "(Selection)" or whatever to TM_DISPLAYNAME the user will be notified additionally. 2) TextMate::Executor.run -> version The interpreter I'm working with is an "hybrid". If you start it without arguments or with arguments which can't be interpreted as a file name the interpreter starts in the GUI mode, i.e. a real application with windows etc. pops up. Otherwise, if you pass as first argument a valid file path plus optional arguments, the interpreter works like Perl, Python, etc. in a command line mode. The current TextMate::Executor.run implementation has a nice feature, you can get the current version number from the current installed interpreter or compiler as default BUT you are NOT able to avoid it. In my case the interpreter doesn't catch these kind of querying, in other words I can set the ":version_*"-options to whatever the GUI is popping up and tells me stories about errors. Finally I ended up to write a special script which returns the version number and I call it via: :version_args => ENV["TM_BUNDLE_SUPPORT"] + "/bin/version.praat" but I believe one could solve this kind of problem quite easily, namely if the developer sets explicitly ":version_args => nil" the version querying won't be executed. To achieve that one has to change one single line: executor.rb line: 99 version = (not options[:version_args].nil?) ? parse_version(args[0], options) : '' Did I miss something? Kind regards, Hans From sanssecours at f-m.fm Mon Sep 29 16:52:12 2014 From: sanssecours at f-m.fm (sanssecours) Date: Mon, 29 Sep 2014 09:52:12 -0700 (PDT) Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: References: Message-ID: <1412009532491-28290.post@n5.nabble.com> Carpii UK wrote > Thanks, I was a paid up BBEdit user before I moved to TM2, but it fell out > of favour after a few bugs (one of them was due to mismatched line endings > funnily enough, it would often misreport line numbers) > > Im able to do it in TM by using Text -> Filter Through Command > > But a specific bundle item would be nice so I could assign a key combo and > bypass the dialog (which often has other commands Ive used in it) > > > > On 29 September 2014 16:37, Matt Neuburg < > matt@ > > wrote: > >> >> On Sep 29, 2014, at 5:26 AM, Carpii UK < > carpii.uk@ > > wrote: >> >> > Is there an existing bundle item which will strip the > > and make the >> full file LF consistent? >> > >> >> I use BBEdit (or its free little brother TextWrangler) for this. With >> BBEdit you can just create a text engine that batch-processes your files. >> m. >> >> -- >> matt neuburg, phd = http://www.apeth.net/matt/ >> pantes anthropoi tou eidenai oregontai phusei >> Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do >> iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do >> RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html >> >> >> >> _______________________________________________ >> textmate mailing list >> > textmate at .macromates >> http://lists.macromates.com/listinfo/textmate >> > > > _______________________________________________ > textmate mailing list > textmate at .macromates > http://lists.macromates.com/listinfo/textmate Hi Carpii and Matt, I just quickly created a bundle item for the task you asked, Carpii. It uses `tr` to delete all carriage returns from your current document. Just press `^??-R` to use the command on the current file. I hope you like it. If not, just leave a reply with your suggestions to make the command better, and I will try to incorporate them. Kind regards, Ren? Carpii.zip -- View this message in context: http://textmate.1073791.n5.nabble.com/Convert-CRLF-to-LF-tp28286p28290.html Sent from the textmate users mailing list archive at Nabble.com. From carpii.uk at gmail.com Mon Sep 29 18:35:03 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Mon, 29 Sep 2014 19:35:03 +0100 Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: <1412009532491-28290.post@n5.nabble.com> References: <1412009532491-28290.post@n5.nabble.com> Message-ID: Hi Ren?, Awesome, thanks so much for taking the time to create this! I didn't realise you could add new menu items with a simple bash scripts like that (Ive looked at bundles in the past but they were hard to decipher). One question.. The bundle works from the saved file (and so any unsaved changes you've made since are not passed to tr). Likewise the bundle cannot run on a unsaved buffer. I wonder if there's a way around that? Its not an issue for your bundle as I can just save it, but would be useful to know for the future Thanks Paul On 29 September 2014 17:52, sanssecours wrote: > Carpii UK wrote > > Thanks, I was a paid up BBEdit user before I moved to TM2, but it fell > out > > of favour after a few bugs (one of them was due to mismatched line > endings > > funnily enough, it would often misreport line numbers) > > > > Im able to do it in TM by using Text -> Filter Through Command > > > > But a specific bundle item would be nice so I could assign a key combo > and > > bypass the dialog (which often has other commands Ive used in it) > > > > > > > > On 29 September 2014 16:37, Matt Neuburg < > > > matt@ > > > > wrote: > > > >> > >> On Sep 29, 2014, at 5:26 AM, Carpii UK < > > > carpii.uk@ > > > > wrote: > >> > >> > Is there an existing bundle item which will strip the > > > > and make the > >> full file LF consistent? > >> > > >> > >> I use BBEdit (or its free little brother TextWrangler) for this. With > >> BBEdit you can just create a text engine that batch-processes your > files. > >> m. > >> > >> -- > >> matt neuburg, phd = http://www.apeth.net/matt/ > >> pantes anthropoi tou eidenai oregontai phusei > >> Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do > >> iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do > >> RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html > >> > >> > >> > >> _______________________________________________ > >> textmate mailing list > >> > > > textmate at .macromates > > >> http://lists.macromates.com/listinfo/textmate > >> > > > > > > _______________________________________________ > > textmate mailing list > > > textmate at .macromates > > > http://lists.macromates.com/listinfo/textmate > > Hi Carpii and Matt, > > I just quickly created a bundle item for the task you asked, Carpii. It > uses > `tr` to delete all carriage returns from your current document. Just press > `^??-R` to use the command on the current file. I hope you like it. If not, > just leave a reply with your suggestions to make the command better, and I > will try to incorporate them. > > Kind regards, > Ren? > > Carpii.zip > > > > -- > View this message in context: > http://textmate.1073791.n5.nabble.com/Convert-CRLF-to-LF-tp28286p28290.html > Sent from the textmate users mailing list archive at Nabble.com. > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From txmt at vanklinkenbergsoftware.nl Mon Sep 29 20:48:19 2014 From: txmt at vanklinkenbergsoftware.nl (Bas Van Klinkenberg) Date: Mon, 29 Sep 2014 22:48:19 +0200 Subject: [TxMt] 'Open Recent' menu regression Message-ID: <590D4A60-FB13-48F5-9BCE-7EA5D268E66F@vanklinkenbergsoftware.nl> Hi, Just wanted to let you know that since version 2.0-alpha.9567 the ?File' -> 'Open Recent...' list has become pretty useless because it now keeps track of each and every file you open inside a project directory, rather than just keeping only the project directory entry in the list. Due to this older items (other projects I worked on yesterday for example) drop off the list very fast, taking the adjective ?Recent? a bit too literal? :-) Best regards, Bas From me at dusek.me Mon Sep 29 22:49:29 2014 From: me at dusek.me (=?utf-8?Q?Boris_Du=C5=A1ek?=) Date: Tue, 30 Sep 2014 00:49:29 +0200 Subject: [TxMt] Cannot build TextMate: test_network and test_SoftwareUpdate fail Message-ID: Hello, building latest textmate after a long time, I am hitting this error: $ ninja TextMate [9/36] Run test ?/Users/dusek/build/TextMate/Frameworks/network/test_network?? FAILED: /Users/dusek/build/TextMate/Frameworks/network/test_network && touch /Users/dusek/build/TextMate/Frameworks/network/test_network.run test_network: 1 of 1 test failed: /Users/dusek/Documents/Projects/Contribute/textmate/Frameworks/network/tests/t_download.cc:58: Expected (network::download(network::request_t(url, &myFilter, NULL), &error) == 200), found (0 != 200) [9/36] Run test ?/Users/dusek/build/TextMate/Frameworks/SoftwareUpdate/test_SoftwareUpdate?? FAILED: /Users/dusek/build/TextMate/Frameworks/SoftwareUpdate/test_SoftwareUpdate && touch /Users/dusek/build/TextMate/Frameworks/SoftwareUpdate/test_SoftwareUpdate.run test_SoftwareUpdate: 5 of 6 tests failed: /Users/dusek/Documents/Projects/Contribute/textmate/Frameworks/SoftwareUpdate/tests/t_sw_update.cc:64: Expected (err == "Extracting archive."), found ("Failed to connect to localhost port 64762: Connection refused" != "Extracting archive.") /Users/dusek/Documents/Projects/Contribute/textmate/Frameworks/SoftwareUpdate/tests/t_sw_update.cc:73: Expected (err == "Extracting archive."), found ("Failed to connect to localhost port 64762: Connection refused" != "Extracting archive.") /Users/dusek/Documents/Projects/Contribute/textmate/Frameworks/SoftwareUpdate/tests/t_sw_update.cc:82: Expected (err == "Bad signature."), found ("Failed to connect to localhost port 64762: Connection refused" != "Bad signature.") /Users/dusek/Documents/Projects/Contribute/textmate/Frameworks/SoftwareUpdate/tests/t_sw_update.cc:91: Expected (err == "Bad signature."), found ("Failed to connect to localhost port 64762: Connection refused" != "Bad signature.") /Users/dusek/Documents/Projects/Contribute/textmate/Frameworks/SoftwareUpdate/tests/t_sw_update.cc:109: Expected (err == NULL_STR), found ("Failed to connect to localhost port 64762: Connection refused" != "?") ninja: build stopped: subcommand failed. I tried ninja network/coerce and ninja SoftwareUpdate/coerce and they do skip the relevant tests, but these choices do not persist and after running ninja TextMate again, I hit the same failures. System: Xcode 6.0.1 (i.e. 10.9 SDK) running on Yosemite DP8 (14A361c), happens even on a totally clean rebuild (git clean -xffd; git reset --hard HEAD; rm -rf ~/build/TextMate/; ./configure && ninja) Is there please anything I can do to make TextMate compile? Thanks, Boris From mailinglist at textmate.org Mon Sep 29 23:07:26 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 30 Sep 2014 01:07:26 +0200 Subject: [TxMt] Re: Cannot build TextMate: test_network and test_SoftwareUpdate fail In-Reply-To: References: Message-ID: <1EBC54A7-93F9-4E47-A2A3-CFFD6FF1FB13@textmate.org> On 30 Sep 2014, at 0:49, Boris Du?ek wrote: > building latest textmate after a long time, I am hitting this error: > [?] Both failures seems to be about not getting through to the test http server TM sets up. Could it be that you have a proxy setup or something else network related that could explain this? > I tried ninja network/coerce and ninja SoftwareUpdate/coerce and they > do skip the relevant tests, but these choices do not persist and after > running ninja TextMate again, I hit the same failures. Unfortunately this no longer work: ninja now references $builddir/.ninja_log to see when something was last succesfully built, so only way to fool it is to edit this file. I need to update the README? > Is there please anything I can do to make TextMate compile? You can delete the two test files (that fail). From saclukey at live.com Mon Sep 29 23:13:02 2014 From: saclukey at live.com (Steven Clukey) Date: Mon, 29 Sep 2014 19:13:02 -0400 Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: References: <1412009532491-28290.post@n5.nabble.com> Message-ID: Paul, You can just use find and replace on \r in regular expression mode. Or control+option+enter will create a CR so you can use that in the find and replace as well. You can run find and replace on the entire project folder to get rid of all of them. Steven On Sep 29, 2014, at 2:35 PM, Carpii UK wrote: > Hi Ren?, > > Awesome, thanks so much for taking the time to create this! > > I didn't realise you could add new menu items with a simple bash scripts like that (Ive looked at bundles in the past but they were hard to decipher). > > One question.. The bundle works from the saved file (and so any unsaved changes you've made since are not passed to tr). > Likewise the bundle cannot run on a unsaved buffer. > > I wonder if there's a way around that? > Its not an issue for your bundle as I can just save it, but would be useful to know for the future > > Thanks > Paul > > > > > > > On 29 September 2014 17:52, sanssecours wrote: > Carpii UK wrote > > Thanks, I was a paid up BBEdit user before I moved to TM2, but it fell out > > of favour after a few bugs (one of them was due to mismatched line endings > > funnily enough, it would often misreport line numbers) > > > > Im able to do it in TM by using Text -> Filter Through Command > > > > But a specific bundle item would be nice so I could assign a key combo and > > bypass the dialog (which often has other commands Ive used in it) > > > > > > > > On 29 September 2014 16:37, Matt Neuburg < > > > matt@ > > > > wrote: > > > >> > >> On Sep 29, 2014, at 5:26 AM, Carpii UK < > > > carpii.uk@ > > > > wrote: > >> > >> > Is there an existing bundle item which will strip the > > > > and make the > >> full file LF consistent? > >> > > >> > >> I use BBEdit (or its free little brother TextWrangler) for this. With > >> BBEdit you can just create a text engine that batch-processes your files. > >> m. > >> > >> -- > >> matt neuburg, phd = http://www.apeth.net/matt/ > >> pantes anthropoi tou eidenai oregontai phusei > >> Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do > >> iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do > >> RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html > >> > >> > >> > >> _______________________________________________ > >> textmate mailing list > >> > > > textmate at .macromates > > >> http://lists.macromates.com/listinfo/textmate > >> > > > > > > _______________________________________________ > > textmate mailing list > > > textmate at .macromates > > > http://lists.macromates.com/listinfo/textmate > > Hi Carpii and Matt, > > I just quickly created a bundle item for the task you asked, Carpii. It uses > `tr` to delete all carriage returns from your current document. Just press > `^??-R` to use the command on the current file. I hope you like it. If not, > just leave a reply with your suggestions to make the command better, and I > will try to incorporate them. > > Kind regards, > Ren? > > Carpii.zip > > > > -- > View this message in context: http://textmate.1073791.n5.nabble.com/Convert-CRLF-to-LF-tp28286p28290.html > Sent from the textmate users mailing list archive at Nabble.com. > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate -------------- next part -------------- An HTML attachment was scrubbed... URL: From carpii.uk at gmail.com Mon Sep 29 23:26:01 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Tue, 30 Sep 2014 00:26:01 +0100 Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: References: <1412009532491-28290.post@n5.nabble.com> Message-ID: Hi Steven, thanks for the reply I do understand I can convert it interactively, I was just looking for a one-shot key combo really. I can also convert them all en-masse via a quick bash loop, but doing this creates a few problems in git, since any historic branches suddenly conflict when I later merge them (and because all the line endings have changed, the whole file is different rather than just blocks of code) For this reason I prefer to just deal with them on a case by case basis as I encounter them Cheers Paul On 30 September 2014 00:13, Steven Clukey wrote: > Paul, > > You can just use find and replace on \r in regular expression mode. Or > control+option+enter will create a CR so you can use that in the find and > replace as well. You can run find and replace on the entire project folder > to get rid of all of them. > > Steven > > > On Sep 29, 2014, at 2:35 PM, Carpii UK wrote: > > Hi Ren?, > > Awesome, thanks so much for taking the time to create this! > > I didn't realise you could add new menu items with a simple bash scripts > like that (Ive looked at bundles in the past but they were hard to > decipher). > > One question.. The bundle works from the saved file (and so any unsaved > changes you've made since are not passed to tr). > Likewise the bundle cannot run on a unsaved buffer. > > I wonder if there's a way around that? > Its not an issue for your bundle as I can just save it, but would be > useful to know for the future > > Thanks > Paul > > > > > > > On 29 September 2014 17:52, sanssecours wrote: > >> Carpii UK wrote >> > Thanks, I was a paid up BBEdit user before I moved to TM2, but it fell >> out >> > of favour after a few bugs (one of them was due to mismatched line >> endings >> > funnily enough, it would often misreport line numbers) >> > >> > Im able to do it in TM by using Text -> Filter Through Command >> > >> > But a specific bundle item would be nice so I could assign a key combo >> and >> > bypass the dialog (which often has other commands Ive used in it) >> > >> > >> > >> > On 29 September 2014 16:37, Matt Neuburg < >> >> > matt@ >> >> > > wrote: >> > >> >> >> >> On Sep 29, 2014, at 5:26 AM, Carpii UK < >> >> > carpii.uk@ >> >> > > wrote: >> >> >> >> > Is there an existing bundle item which will strip the >> > >> > and make the >> >> full file LF consistent? >> >> > >> >> >> >> I use BBEdit (or its free little brother TextWrangler) for this. With >> >> BBEdit you can just create a text engine that batch-processes your >> files. >> >> m. >> >> >> >> -- >> >> matt neuburg, phd = http://www.apeth.net/matt/ >> >> pantes anthropoi tou eidenai oregontai phusei >> >> Programming iOS 7! http://shop.oreilly.com/product/0636920031017.do >> >> iOS 7 Fundamentals! http://shop.oreilly.com/product/0636920032465.do >> >> RubyFrontier! http://www.apeth.com/RubyFrontierDocs/default.html >> >> >> >> >> >> >> >> _______________________________________________ >> >> textmate mailing list >> >> >> >> > textmate at .macromates >> >> >> http://lists.macromates.com/listinfo/textmate >> >> >> > >> > >> > _______________________________________________ >> > textmate mailing list >> >> > textmate at .macromates >> >> > http://lists.macromates.com/listinfo/textmate >> >> Hi Carpii and Matt, >> >> I just quickly created a bundle item for the task you asked, Carpii. It >> uses >> `tr` to delete all carriage returns from your current document. Just press >> `^??-R` to use the command on the current file. I hope you like it. If >> not, >> just leave a reply with your suggestions to make the command better, and I >> will try to incorporate them. >> >> Kind regards, >> Ren? >> >> Carpii.zip >> >> >> >> -- >> View this message in context: >> http://textmate.1073791.n5.nabble.com/Convert-CRLF-to-LF-tp28286p28290.html >> Sent from the textmate users mailing list archive at Nabble.com. >> >> _______________________________________________ >> textmate mailing list >> textmate at lists.macromates.com >> http://lists.macromates.com/listinfo/textmate >> > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > > > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From mailinglist at textmate.org Mon Sep 29 23:38:16 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 30 Sep 2014 01:38:16 +0200 Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: References: <1412009532491-28290.post@n5.nabble.com> Message-ID: On 30 Sep 2014, at 1:26, Carpii UK wrote: > I do understand I can convert it interactively, I was just looking for > a > one-shot key combo really. Edit ? Macros ? Start Recording ?F, search for \r, replace with empty string and click Replace All. Edit ? Macros ? Stop Recording Edit ? Macros ? Save Macro Now you have a macro for this to which you can assign a key equivalent or call up with ??T. From carpii.uk at gmail.com Tue Sep 30 00:08:50 2014 From: carpii.uk at gmail.com (Carpii UK) Date: Tue, 30 Sep 2014 01:08:50 +0100 Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: References: <1412009532491-28290.post@n5.nabble.com> Message-ID: Very cool, thanks Allan Starting to realise I don't know TM as well as I thought I did :-) On 30 September 2014 00:38, Allan Odgaard wrote: > On 30 Sep 2014, at 1:26, Carpii UK wrote: > > I do understand I can convert it interactively, I was just looking for a >> one-shot key combo really. >> > > Edit ? Macros ? Start Recording > ?F, search for \r, replace with empty string and click Replace All. > Edit ? Macros ? Stop Recording > Edit ? Macros ? Save Macro > > Now you have a macro for this to which you can assign a key equivalent or > call up with ??T. > > > > _______________________________________________ > textmate mailing list > textmate at lists.macromates.com > http://lists.macromates.com/listinfo/textmate > -------------- next part -------------- An HTML attachment was scrubbed... URL: From me at dusek.me Tue Sep 30 07:31:00 2014 From: me at dusek.me (=?utf-8?Q?Boris_Du=C5=A1ek?=) Date: Tue, 30 Sep 2014 09:31:00 +0200 Subject: [TxMt] Re: Cannot build TextMate: test_network and test_SoftwareUpdate fail In-Reply-To: <1EBC54A7-93F9-4E47-A2A3-CFFD6FF1FB13@textmate.org> References: <1EBC54A7-93F9-4E47-A2A3-CFFD6FF1FB13@textmate.org> Message-ID: <941F95F1-F63F-4501-A6D6-D7F7B7FA421B@dusek.me> > Both failures seems to be about not getting through to the test http server TM sets up. > > Could it be that you have a proxy setup or something else network related that could explain this? I am not aware about having any non-standard setup, I don?t think I ever did anything else network-wise than connecting to a Wi-Fi network. I tried again this morning (to try if it could be some temporary issue with my internet provider), I also tried connecting to internet through my iPhone as a hot-spot, I still get the same errors in both tests. I have firewall off in System Preferences > Security & Privacy. In System Preferences > Network > Wi-Fi > Advanced? > Proxies > Select a protocol to configure:, I have no protocol checked. >> I tried ninja network/coerce and ninja SoftwareUpdate/coerce and they do skip the relevant tests, but these choices do not persist and after running ninja TextMate again, I hit the same failures. > > Unfortunately this no longer work: ninja now references $builddir/.ninja_log to see when something was last succesfully built, so only way to fool it is to edit this file. I need to update the README? Looking forward to the updated README, as I cannot make sense of how to edit .ninja_log to achieve what I want. >> Is there please anything I can do to make TextMate compile? > > You can delete the two test files (that fail). This does not help as on subsequent ninja call, the files get reinstantiated and the same tests fail again. Thanks for the suggestions so far. From sanssecours at f-m.fm Tue Sep 30 09:59:27 2014 From: sanssecours at f-m.fm (sanssecours) Date: Tue, 30 Sep 2014 02:59:27 -0700 (PDT) Subject: [TxMt] Re: Convert CRLF to LF In-Reply-To: References: <1412009532491-28290.post@n5.nabble.com> Message-ID: <1412071167532-28300.post@n5.nabble.com> Hi Paul, Carpii UK wrote > Hi Ren?, > > Awesome, thanks so much for taking the time to create this! > > I didn't realise you could add new menu items with a simple bash scripts > like that (Ive looked at bundles in the past but they were hard to > decipher). > > One question.. The bundle works from the saved file (and so any unsaved > changes you've made since are not passed to tr). > Likewise the bundle cannot run on a unsaved buffer. > I wonder if there's a way around that? Allan already mentioned one possibility. Another one would be to use the current document as input for the command. I updated the bundle command accordingly. It now uses a small piece of Python code to remove all carriage return symbols from the current document. Carpii UK wrote > Its not an issue for your bundle as I can just save it, but would be > useful > to know for the future > > Thanks > Paul Kind regards, Ren? Carpii.zip -- View this message in context: http://textmate.1073791.n5.nabble.com/Convert-CRLF-to-LF-tp28286p28300.html Sent from the textmate users mailing list archive at Nabble.com. From mailinglist at textmate.org Tue Sep 30 12:22:45 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 30 Sep 2014 14:22:45 +0200 Subject: [TxMt] Re: Cannot build TextMate: test_network and test_SoftwareUpdate fail In-Reply-To: <941F95F1-F63F-4501-A6D6-D7F7B7FA421B@dusek.me> References: <1EBC54A7-93F9-4E47-A2A3-CFFD6FF1FB13@textmate.org> <941F95F1-F63F-4501-A6D6-D7F7B7FA421B@dusek.me> Message-ID: On 30 Sep 2014, at 9:31, Boris Du?ek wrote: >> Both failures seems to be about not getting through to the test http >> server TM sets up. >> >> Could it be that you have a proxy setup or something else network >> related that could explain this? > > I am not aware about having any non-standard setup, I don?t think I > ever did anything else network-wise than connecting to a Wi-Fi > network. > > I tried again this morning (to try if it could be some temporary issue > with my internet provider), I also tried connecting to internet > through my iPhone as a hot-spot, I still get the same errors in both > tests. I mentioned the proxy because TextMate reads the proxy settings but looking at the code, it does skip that for localhost. And this is code connecting to localhost so internet environment shouldn?t matter. Re-reading the error you get, it does say: Failed to connect to localhost port 64762: Connection refused. So the issue seems to be that the server that the test code sets up fails to listen on this port. The server setup does abort() if it fails to create the listening socket, but it doesn?t report any errors if accept() fails, so I added a message for that and pushed. Though this is grasping at straws? Have you rebooted since you got this error? Again, grasping at straws, but I have never seen this code fail before and it?s about setting up a listening socket? > [?] > Looking forward to the updated README, as I cannot make sense of how > to edit .ninja_log to achieve what I want. Afraid my edit will just remove all info about this. There isn?t any practical solution to skipping tests and I will likely have to make it a ./configure argument. >> You can delete the two test files (that fail). > This does not help as on subsequent ninja call, the files get > reinstantiated and the same tests fail again. I think you misunderstand. You need to delete Frameworks/network/tests/t_download.cc ? there is no way ninja would restore that. From mailinglist at textmate.org Tue Sep 30 13:37:06 2014 From: mailinglist at textmate.org (Allan Odgaard) Date: Tue, 30 Sep 2014 15:37:06 +0200 Subject: [TxMt] Re: 'Open Recent' menu regression In-Reply-To: <590D4A60-FB13-48F5-9BCE-7EA5D268E66F@vanklinkenbergsoftware.nl> References: <590D4A60-FB13-48F5-9BCE-7EA5D268E66F@vanklinkenbergsoftware.nl> Message-ID: <5F1EEE57-5B7D-421C-B12B-80FF40382540@textmate.org> On 29 Sep 2014, at 22:48, Bas Van Klinkenberg wrote: > Just wanted to let you know that since version 2.0-alpha.9567 the > ?File' -> 'Open Recent...' list has become pretty useless because it > now keeps track of each and every file you open inside a project > directory [?] Are you referring to this change as the regression: * Files opened via `mate` are now added to the recent menu by default unless the file is a dot-file (hidden), `--wait` or `--no-recent` is specified, or the file is in the system?s temporary folder. Also cleaned up the command options a bit, e.g. `--async` is now `--no-wait`, run `mate -h` for more info. > Due to this older items (other projects I worked on yesterday for > example) drop off the list very fast There is a special action for recent ?projects? accessible as the misnamed Open ? Favorites? (??O). -------------- next part -------------- An HTML attachment was scrubbed... URL: From samson.etienne at gmail.com Tue Sep 30 15:10:24 2014 From: samson.etienne at gmail.com (Etienne Samson) Date: Tue, 30 Sep 2014 17:10:24 +0200 Subject: [TxMt] Re: 'Open Recent' menu regression In-Reply-To: <5F1EEE57-5B7D-421C-B12B-80FF40382540@textmate.org> References: <590D4A60-FB13-48F5-9BCE-7EA5D268E66F@vanklinkenbergsoftware.nl> <5F1EEE57-5B7D-421C-B12B-80FF40382540@textmate.org> Message-ID: <866314DC-C520-4F43-8972-3D74D68C81A2@gmail.com> Le 30 sept. 2014 ? 15:37, Allan Odgaard a ?crit : > On 29 Sep 2014, at 22:48, Bas Van Klinkenberg wrote: > > Just wanted to let you know that since version 2.0-alpha.9567 the ?File' -> 'Open Recent...' list has become pretty useless because it now keeps track of each and every file you open inside a project directory [?] > > Are you referring to this change as the regression: > > ? Files opened via mate are now added to the recent menu by default unless the file is a dot-file (hidden), --wait or --no-recent is specified, or the file is in the system?s temporary folder. Also cleaned up the command options a bit, e.g. --async is now --no-wait, run mate -h for more info. > Due to this older items (other projects I worked on yesterday for example) drop off the list very fast > > There is a special action for recent ?projects? accessible as the misnamed Open ? Favorites? (??O). Hey, I have been wondering about that, though it doesn't really match OP's description (keeps track of every file you open), but for me it really keep tracks of every file you *create*. Here are some steps : - Open one of your projects with Terminal and mate, or the Recent menu itself ? doesn't see to matter. - Create and save a new empty file. - A new entry for the file is added to the Recents menu, even thought it's under a known project. I wouldn't have expected this new file to be added under the Recent menu, since it's already "accessible" from a project I use day-to-day, and I don't think it falls under your changes made to mate. Cordialement, Etienne Samson -- samson.etienne at gmail.com From me at dusek.me Tue Sep 30 20:28:55 2014 From: me at dusek.me (=?utf-8?Q?Boris_Du=C5=A1ek?=) Date: Tue, 30 Sep 2014 22:28:55 +0200 Subject: [TxMt] Re: Cannot build TextMate: test_network and test_SoftwareUpdate fail In-Reply-To: References: <1EBC54A7-93F9-4E47-A2A3-CFFD6FF1FB13@textmate.org> <941F95F1-F63F-4501-A6D6-D7F7B7FA421B@dusek.me> Message-ID: <34ADFF1A-7DBA-43B8-9BE4-CE5B527F564D@dusek.me> > The server setup does abort() if it fails to create the listening socket, but it doesn?t report any errors if accept() fails, so I added a message for that and pushed. Though this is grasping at straws? > > Have you rebooted since you got this error? Again, grasping at straws, but I have never seen this code fail before and it?s about setting up a listening socket? Running at latest master (639f1d282f381ea227098566cdea96c98fd3f653) and after reboot, a complete clean rebuild. I still get the errors as before, and I get no new error messages that you added in 7cb2330aaef49e3a83ed8147cf2a3e74e91a8e1d. >>> You can delete the two test files (that fail). >> This does not help as on subsequent ninja call, the files get reinstantiated and the same tests fail again. > > I think you misunderstand. You need to delete Frameworks/network/tests/t_download.cc ? there is no way ninja would restore that. You are right, I tried removing the test_network and test_network.run files before. Removing the relevant .cc files enabled me to finally compile and launch TextMate :-)