-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512
Hello,
the update to the PHP-Syntax for @uses is not working. The correct
line for the match is this:
match = '\@(a(ccess|uthor)|c(ategory|opyright)|global|li(cense|nk)|pa
(ckage|ram)|return|s(ee|ince|tatic|ubpackage)|t(hrows|odo)|v(ar|
ersion)|uses)\b';
The last brace was closed to early and didn't include the "uses", so
the @ of uses isn't matched. Please fix this.
Thanks in advance,
Simon Ruderich
- ----
> privacy is necessary
> using http://gnupg.org
> public key id: 0x6115F804EFB33229 http://ruderich.com/
simonruderich.asc
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.5 (Darwin)
iD8DBQFFPoEaYRX4BO+zMikRCgMsAKCpeBhNvp9TaK1BDsEhdzdSGMA8hwCeIOwW
4VzxJ6KkWmA3K9NaDhVqF0U=
=tgnn
-----END PGP SIGNATURE-----
Hi, perhaps I haven't understood the syntax correctly, but I have
tried editing the LaTeX configuration file and I am getting no love.
I wanted to add a quote environment. I tried the following which has
no apparent effect:
quote = {
triggers = ( 'qu', 'quote');
};
All the best, Mark
_________________
Mark Eli Kalderon
Department of Philosophy
University College London
Gower Street
London WC1E 6BT
Dept webpage: http://www.ucl.ac.uk/philosophy
Personal wepage: http://www.kalderon.demon.co.uk
Hello!
I found an earlier post in the archiv files, but it seems, that this
command isn´t realised yet.
My Latex projects contain multiple files and the Code Brwoser only
shows the outline of the active one.
What I need, is an outline of the whole project. Are there any
solutions?
Helge
Is it possible to provide a list of options to someone in a snippet?
For example at a tabstop, rather than them writing in text, you could
provide a list of options that they could select from?
I'm imagining something similar to the one that pops up to
disambiguate shortcuts.
Ed
Is it possible to have a repeating tabstop in a snippet?
for example it would be great to have a snippet to insert a html option list:
<select>
<option value="$1">$2</option>
</select>
You tab to $1, fill in the value, tab to $2, fill in the content, then
when you press tab again, it inserts another option:
<select>
<option value="1">Cherry</option>
<option value="$3">$4</option>
</select>
And carries on doing so for an arbitrarily long time.
If it's possible to do this I'd be very grateful.
Thanks
Ed
On 24.10.2006, at 13:27, textmate-request(a)lists.macromates.com wrote:
> Heiko Witte schrieb:
>> Hi,
>>
>> every mail from this mailing list is signed with a PGP key
>> (0x7B7C623764AB61E7). Where can I get this public key ? I already
>> searched the mailing lists but couldn't find any related topic.
>>
>> Thanks
>>
>> Heiko
>
> Every email? I just searched pgp.mit.edu for this key and came up with
> my own key ;)
>
> http://pgp.mit.edu:11371/pks/lookup?search=0x7B7C623764AB61E7&op=index
>
> Regards, Horst
I added your key to my gpg keychain, now everything's fine =)
Thanks, Heiko
Hi,
every mail from this mailing list is signed with a PGP key
(0x7B7C623764AB61E7). Where can I get this public key ? I already
searched the mailing lists but couldn't find any related topic.
Thanks
Heiko
FYI for LaTeX bundle maintainers and users:
I recently updated XeTeX (to v0.995, 16-Aug) and discovered that
previewing in TextMate had become kind of cranky (though running
xelatex from the command line still worked fine).
The crankiness: using "Typeset & View (PDF)", the first page of
output would be the following, in Computer Modern:
[zf ]preparseRendererAAT,ICU
The correct LaTeX output would appear starting on the next page.
After a heap of digging around (resulting in [1], which pointed to a
path problem with xkeyval.tex), I figured out that the call to
kpsewhich is causing the wrong paths to get expanded (in particular,
the texmf.tetex versions were taking precedence over texmf.local).
This seems to fix it for me:
Old: export TEXINPUTS="$TM_BUNDLE_SUPPORT/tex//:$(kpsewhich --expand-
var '$TEXINPUTS')"
New: export TEXINPUTS="$TM_BUNDLE_SUPPORT/tex//:$(kpsewhich -progname
$TEX --expand-var '$TEXINPUTS')"
HTH, as they used to say.
--ds
[1] http://www.tug.org/pipermail/xetex/2006-August/004700.html
To add on to James' nifty drag commands, here's mine for adding a
complete background line in a definition. It's tab stopped a little
differently than the normal background command, as I have modified
mine to the way that my brain works.
I'm not sure how the rest of you will feel about the newline at the
end, it works for me with the way I format my code. Feel free to
pull it and the -e argument out from the echo statement.
File Types: png, jpeg, jpg, gif
echo -ne "background:\${1: #\${2:DDD}} url('$TM_DROPPED_FILE')\${3: \$
{4:repeat/repeat-x/repeat-y/no-repeat} \${5:scroll/fixed} \${6:top/
center/bottom/x-%/x-pos} \${7:left/center/right/y-%/y-pos}};\n\$0"
Scope: meta.property-list.css
Brett Terpstra : Art Director
Circle Six Design, Inc.
111 Riverfront Dr, Suite 204
..................................................
p: 507.459.4398
877.858.4332
f: 1.866.540.3063
e: brett(a)circlesixdesign.com
http://www.circlesixdesign.com
..................................................

Brett Terpstra : Art Director
Circle Six Design, Inc.
111 Riverfront Dr, Suite 204
..................................................
p: 507.459.4398
877.858.4332
f: 1.866.540.3063
e: brett(a)circlesixdesign.com
http://www.circlesixdesign.com
..................................................
Is there a way to scope a snippet in the CSS bundle to mimic the
Special: Return Inside Empty Open/Close Tags in the HTML bundle? I'd
like to have the same behavior between empty brackets, but they
obviously behave differently and I have a very vague understanding of
the language file at this point. I got the command to work the first
time, but now it adds an extra line break every time I hit enter.
Any pointers?
Thanks,
Brett
Brett Terpstra : Art Director
Circle Six Design, Inc.
111 Riverfront Dr, Suite 204
..................................................
p: 507.459.4398
877.858.4332
f: 1.866.540.3063
e: brett(a)circlesixdesign.com
http://www.circlesixdesign.com
..................................................
Not sure if anyone has done these yet but here are some simple drag
actions for images in CSS.
The second one if for dropping between url() so you can use the
background snippet for instance then drag an image between the brackets.
Image as a background:
File types:
png, jpg, jpeg, gif
Commands:
echo "background-image: url(\"$TM_DROPPED_FILE\");"
Scope:
meta.property-list.css
Image Path:
File types:
png, jpg, jpeg, gif
Commands
echo -n "\"$TM_DROPPED_FILE\""
meta.property-value.css
I don't really know how to ask this question in a simple manner so here
it goes:
I'm doing some HTML coding and need to add JavaScript using the
Prototype framework. In this early develpment phase I like to write
JavaScript embedded in HTML. When I write a script tag,
"source.js.embedded.html" gets added to the scope. How can I tell
TextMate to use features from the Prototype & Scriptaculous bundle in
addition to the JavaScript default one? Shifting languages is tedious
and messes up the highlighting.
I have a problem with Brett's "Wrap Text as Link with Title". It
works fine, also gets the link from the web, but then something like
this is inserted:
2006-10-23 13:38:42.367 CocoaDialog[569] InputManagersManager startup
<a href="http://macromates.com/" title="TextMate — The Missing Editor
for Mac OS X" >TextMate</a>
the second line is perfect, but i wonder where the first line came
from? i talked to brett about it but he is not able to reproduce the
error. any ideas?
thanks!
Tilman.
_______________________________________________________________
Tilman.
net citizen /// designer /// teacher
____ digital homes ::: personal http://www.tint.de ///
collective http://www.dkdl.de
____ analogue homes ::: germany /// nuernberg /// lauf ///
heuchling
____ mailto ::: tilman(a)tint.de
::: the future is in beta.
when I attempt to compile (read: typeset and view) a LaTeX document I
get the following errors:
/bin/bash: line 34: require_cmd: command not found /bin/bash: line 45:
latexErrWarnHtml.py: command not found Click Here to Preview
I am unable to figure out which script is responsible for these
errors. If anybody has had simmilar problems how did you fix them, or
if you know where these might be coming from then I could start
looking there.
Oh, and I am using the i-installer version of tex.
Further more I would like to propose using '1>&2' in scripts to pipe
output to the console where it can be more easily viewed for debugging
purposes.
-adam
Adam Hitchcock Macintosh software at InfImp.net No trees were harmed
in the sending of this message, however,
otaking(a)mac.com | cell - 734.769.0085 | aim - AlfredHi | pgp keys a
great number of electrons were severely inconvenienced.
The Latex file I am working on is getting so long that formatting is
taking a full minute, even for small changes. I realize I could
input chapters as separate files and not load them all each time, but
I prefer not to do this for navigation purposes. I saw on Haris'
blog, "I was editing an html file, and I used the internal html
preview window that it has, and then I left the window open. The
miracle that occurred is that as I continued editing, the preview
window was automatically updated instantly! Without me even having to
save the file!" This does not seem to work for the Latex bundle.
But even if it did, the TM internal preview does not respond to
pdfsync. So my wishful thoughts are for a "flash mode" for the
Latex bundle that is compatible with pdfsync. So many dreams are
coming true with all this good energy within this list, I thought I
would ask...
Jenny
Just a small change to the regex to handle two part url's (i.e.
site.com) which I had inadvertently ignored previously. Also limits
the domain to 2-3 characters because it now recognizes anything
joined by a dot (conjoined.twins). I thought this was easier than
trying to list every possible domain in the regex.
Dear all,
maybe you know the bundle 'Find All in Document'. This script lists
all occurrences for a specific search pattern as an HTML page. You
can click on a line and you jump to that line in the document.
I rewrote the script and I renamed it to 'Hypersearch'. (see attachment)
Changes/Improvements:
- case insensitive search
- result is shown as a table with line numbers (i.e. a better line
distinction)
- search result within a line is highlighted using bold and italics
- there is no need to save your file in beforehand
- it is 'relatively' easy to change the code
- the code is 'rather' unicode save (i.e. It is not a problem to find
a Japanese string. Only some regular expressions are not supported yet.)
- tabs and spaces are shown like the source code (tabs are replaced
by four spaces)
- < and > are shown correctly
- displays the number of lines which match against the search pattern
Problems:
- I tried to implement to jump not only to the line but also to the
right column but unfortunately it works only for ASCII. If there is
e.g. a utf-8 'ö' or a Japanese character within the line the column
number didn't fit. I used perl's pos() function but I get only the
numbers of bytes. Hints are welcomed!!
- Case insensitive search works but not for accented characters like
ç, ä, é etc. You have to use e.g. the pattern 'ç|Ç' instead.
- You cannot search for '§¢' and '¢§'. These strings are used
internally.
The access key is ctrl+shift+f.
Implementation is done with CocoaDialog, grep, sed, and a bit perl
within bash.
If someone find some bugs or have improvements etc., please don't
hesitate to write me!
In other words, COMMENTS ARE WELCOMED!
Best,
Hans-Joerg Bibiko
Okay, this is my version of the wrap text as link command. It seems
to be failing elegantly now, although I need some testing on other
installations as there are some inconsistencies between my two machines.
It recognizes a link in the clipboard, with or without the leading
protocol, and grabs the title for the page off the net to insert as
the title of the link. If there is no protocol on the link it will
attempt to use http://. If the link is bogus or unreachable it will
insert the link that it found but put "Unknown" for the title. Both
the link and the title are tab-stopped. I have it set up under CTRL-
SHIFT-L so that you can choose between the original and the link with
title. Even if you cut out the title grabbing part, it's still seems
like a better way to recognize a link in the clipboard and utilize it.
Brett
Hi!
I just was interested how you guys create screencasts. I looked a
little bit around in the net but could only find Snapz X Pro which
could do the task. Any cheap or free (as in beer) possibilities?
Would have needed it the last days for making some instructions.
Niels
When I start TextMate, it downloads the latest update. I click on
Install & Relaunch and get an "Install Failed" message. This has
happened every day this week.
I searched the archives and found a message that said this happens on
non-admin accounts, but without a fix.
I am an admin account, so something else must be going on.
Anyone else seeing this or any advice?
Bob
Is it possible to define a shell variable by referencing another shell
variable? Here's an example of what I am trying to do:
(These are project variables)
TM_FLEX_ROOT=$TM_PROJECT_DIRECTORY/flex
TM_SERVICES=$FLEX_ROOT/config/WEB-INF/flex/services-config.xml
Thanks
Todd
Hello,
I have been playing with the existing Erlang bundle to improve it a bit
and I have trouble adding code folding.
Here is a simple Erlang function:
start(Var) ->
test(Var),
%% Comment
ok.
The last dot marks the end of the function.
I know that the end of the function is not at the same indent level that
the start marker, but this is always the case in Erlang.
I try to match the end marker, with any number of spaces / tabs at the
beginning of the line but at best, start and stop marks are put in place
but not foldable.
I suppose this is because Textmate does some checks on the number of
spaces or tabs but is there a way to ignore this match ?
If not, is there an estimated release date for this feature ?
Thank you !
--
Mickaël Rémond
http://www.process-one.net/
As a recovering vimaholic, I miss special highlighting of (FIXME|TODO|
XXX) in my source. (I realize that I can get a list of TODOs with C-
S-t, but it's nice to have things like "XXX: broken" or "FIXME
dsandler:" leap out at me as I'm skimming code.)
I seem to be able to hack these rules into individual languages; for
example, in Python, I changed the comment.line.number-sign.py pattern
thusly:
{ name = 'comment.line.number-sign.py';
begin = '#';
end = '$';
patterns = (
{ name = 'todo.todo';
match = '(TODO)( \S*:|:?)';
},
{ name = 'todo.fixme';
match = '(FIXME)( \S*:|:?)';
},
{ name = 'todo.xxx';
match = '(XXX)( \S*:|:?)';
},
{ name = 'todo.changed';
match = '(CHANGED)( \S*:|:?)';
},
);
},
(This seemed to require changing Python's line-comment pattern from a
'match' to a begin/end kind of thing.)
With these scopes, you can now set all TODOs to a given color, or
pick different colors for each, etc. (perhaps matching the colors up
with those in the "Show TODO List" command).
So, is there any way to apply this kind of new scope to to (comments
in) all languages at once, or do I (or the TextMate BDFL) have to go
and integrate it into each one by hand?
Thanks,
--ds
Hello everyone,
The short story is that running ARD (Apple Remote Desktop) Client is
the culprit that is stealing the key sequence.
I don't know why ARD steals the key sequence, but turning it off
(System Preference - under Sharing) eliminates this problem for me.
I'd like to run the ARD client, so if anyone knows how to have the
ARD client run without it stealing the crtl-esc key sequence, I'd
like to hear about it.
I hope this ends the problem for others as well.
peace,
icb
Ira Cary Blanco
Marketing & Media
Samson Technologies Corp.
Samson | Hartke | Zoom
voice: 631.784.2200 x107
fax: 631.784.2201
email: blanco(a)samsontech.com
web: samsontech.com
Hi,
I've been using TM for quite some time now and it's a great editor.
The last weeks I had to write some code for a site using MODx
(www.modxcms.com)
MODx uses some special tags:
[[Snippet]]
[!Noncaching-Snippet!]
{{Chunk}}
[~link~]
[*attribute*]
I added some code to the HTML-Bundle for recognizing this which works
quite good.
modx = { patterns = (
{ name = 'meta.tag.inline.modx.snippet';
begin = '(\[\[)(\w*)(.*?)';
end = '(\]\])';
captures =
{ 1 = { name = 'punctuation.modx.snippet'; };
2 = { name = 'entity.name.tag.modx.snippet'; };
};
},
{ name = 'meta.tag.inline.modx.snippet.nocache';
begin = '(\[\!)(\w*)(.*?)';
end = '(\!\])';
captures =
{ 1 = { name = 'punctuation.modx.nocachesnippet'; };
2 = { name =
'entity.name.tag.modx.snippet.nocache'; };
};
},
{ name = 'meta.tag.inline.modx.link';
begin = '(\[\~)(\w*)';
end = '\~\]';
captures =
{ 1 = { name = 'punctuation.modx.link'; };
2 = { name =
'entity.name.tag.modx.snippet.nocache'; };
};
},
{ name = 'meta.tag.inline.modx.setting';
begin = '(\[\()(\w*)';
end = '(\)\])';
captures =
{ 1 = { name = 'punctuation.modx.setting'; };
2 = { name = 'entity.name.tag.modx.setting'; };
};
},
{ name = 'meta.tag.inline.modx.chunk';
begin = '({{)(\w*)';
end = '(}})';
captures =
{ 1 = { name = 'punctuation.modx.chunk'; };
2 = { name = 'entity.name.tag.modx.chunk'; };
};
},
{ name = 'meta.tag.inline.modx.attribute';
begin = '(\[\*)(\w*)';
end = '(\*\])';
captures =
{ 1 = { name = 'punctuation.modx.attribute'; };
2 = { name = 'entity.name.tag.modx.attribute'; };
};
},
{ name = 'variable.parameter.modx';
match = '(&\w+)=';
},
{ name = 'string.quoted.modx';
begin = '`';
end = '`';
},
{ name = 'invalid.illegal.modx.bad-snippetcall';
match = '(\[[\[\!])(\w+)(\s+)(.+?)([\!\]]\])';
},
);
};
Now my problem is:
There are some Snippet-calls which take parameters:
[[Snippet? ¶meter=`value` ¶meter2=`value2`]]
I tried to define the parameters as variable.parameter.modx and the
values as string.quoted.modx but TM doesn't seem to recognize this.
Also when I write a MODx Tag into a html-string (e.g. <a
href="[(base_url)]">...) TM will ignore it.
Last I tried to highlight invalid snippet calls (e.g. [[Snippet
¶m... (missing ?) or [[Snippet? ¶m='foo' (normal quotation
mark)) but this won't work either.
Can you tell me what I did wrong? :)
Regards
Max Lohrmann
I'd be super-happy if someone could help me with the following
Textmate behaviour.
a. The cursor is in any line that has more characters than the
previous line, like so:
line 1 | ¬
line 2 | 12345¬
When the cursor is at the end (column 6) of line 2 and moved up one
line then it stays in column 6, i.e. behind the ¬ end-of-line marker
of line 1. Like this:
line 1 | ¬ x
line 2 | 12345¬
b. When now return/enter is hit so that a new line gets inserted then
there are also spaces inserted (filling up the place between the end-
of-line marker and the previous place of the cursor) in line 1.
line 1 | ¬
line 2 | ¬
line 3 | 12345¬
Is there any reasoning behind this? It feels really strange to me and
conflicts with several editing habits of mine.
While editing code the insertion of whitespace in line 1 actually
leads to masses of whitespace cluttered all over the place. I can't
imagine that this is intentional. :)
Any hints?
--
sven fuchs fon: +49 (58 45) 98 89 58
artweb design fax: +49 (58 45) 98 89 57
breite straße 65 www: http://www.artweb-design.de
de-29468 bergen mail: svenfuchs(a)artweb-design.de
--
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.
(This isn't specifically a Textmate question, but I know a lot of Textmate
users use Markdown, so I was hoping someone could help.)
How do I name Markdown files so that they are:
1) Recognized automatically by Textmate as Markdown files when you open
them,
2) Indexed automatically by Spotlight,
3) Use a unique extension (like .markdown) rather than a generic .txt
extension?
Right now, if I use the default .markdown extension, the file isn't indexed
by Spotlight. If I save a Markdown file without any extension, it's neither
indexed by Spotlight, nor is it recognized by Textmate as a Markdown file.
A .txt extension is recognized by Textmate and indexed by Spotlight, but I
was hoping to use a different file extension--one that could distinguish
Markdown files from other text files.
How do other people solve this problem?
Thanks,
Tom
**
Okay, I'm trying to keep this one simple and meet my own needs at the
same time. First of all, the insert color command didn't recognize
abbreviated (3 character) RGB color codes. This had 2 drawbacks:
The tab stop inserted by any of the CSS commands, when used in
combination with Insert Color, would add an extra # sign and it
wouldn't properly set the default color when it loaded the color
picker. This revised command is my usual unskilled kludge, but it
recognizes both 3 and 6 character rgb strings, strips the # sign if
unnecessary, and if the resulting hex code is a series of double
repeating characters it will abbreviate (BBAADD becomes BAD). I
wanted to use squeeze to do this in Ruby, but had trouble with colors
like 0000FF, where it would just give back 0F. So I made a mess of
it. It works, but I'd really like to see it cleaned up (I know Haris
could do this in about 4 lines of code). So if anyone has the
time... It'd be great to learn a new algorithm to handle this.
As a side note, I've been using Color Schemer Studio to create
websmart palettes. I can drag directly from the current color
selection into textmate, but it was less convenient than the color
picker. So I've been creating the palettes, and then importing them
into the color picker. That's where this command came from...
Thanks,
Brett
I've become quite a fan of the Rspec BDD framework, and so I whipped
up a bunch of snippets to make my life easier. If anyone wants them,
they're here:
http://blog.amber.org/2006/10/16/textmate-bundle-for-rspec/
Chris
--
| Christopher Petrilli
| petrilli(a)gmail.com
Suppose I am in the html grammar with a php block, like:
<?php
echo 'my insertion point is here';
?>
my expectation is to have ctrl-return produce <br /> rather than the
\n given that I am within the '".
am I wrong on this ?
--
dc
-----
David Clark
Web Specialist
Institute for Community Inclusion (http://www.communityinclusion.org/)
david.clark(a)umb.edu
(617) 287-4318
For a few days now, completion with esc has not been working.
What could be the cause?
Conflict with other modules? or what?
It does not work with ref{} or cite{}.
Any suggestions appreciated.
Christopher Brewster
*****************************************************
Natural Language Processing Group,
Department of Computer Science, University of Sheffield
Awesome, apreciate it Allan.
Very impressed you're on here answering questions. Consider me a an
"offical" textmate user now!
-J
> From: Allan Odgaard <throw-away-1(a)macromates.com>
> Date: October 20, 2006 1:42:15 AM EDT
> To: TextMate users <textmate(a)lists.macromates.com>
> Subject: Re: [TxMt] Executing command on all files in a project AND
> using TM variables in a search and replace
> Reply-To: TextMate users <textmate(a)lists.macromates.com>
>
>
> On 19. Oct 2006, at 22:12, J Fishwick wrote:
>
>> Command:
>> eval arr=("$TM_SELECTED_FILES")
>> for (( i = 0; i < ${#arr[@]}; i++ )); do
>> perl -pe 'while ($string =~ m/mc\d+(.*\n.*True)/g) {$string =~
>> s/mc\d+/tf$TM_FILENAME/g;}';
>> done
>>
>> Input: Entire Doc
>> Output: Replace selected text
>
> You need to set input/output to none/discard (or show tool tip, or
> maybe show as HTML and make your command output some progress info).
>
> Then your actual command needs to load the file itself, it already
> iterators over the selected files, it just doesn’t tell perl to
> load/overwrite each of them.
>
> So a simple example would instead be:
>
> eval arr=("$TM_SELECTED_FILES")
> for f in "${arr[@]}"; do
> perl -i.bak -pe 's/bar/bar/' "$f"
> done
> rescan_project # do this after you make changes to files, so TM
> will rescan
>
> This will replace foo with bar in all selected files (and create
> a .bak file with the old contents).
>
> Two other things: 1) inside single quoted strings, shell variables
> are not expanded, so if you want to use $TM_… then you need to use
> double quotes, and 2) TM_FILENAME is the filename of the file open
> when you called the file, it won’t update inside the loop. Here you
> would instead use the current file given by the loop (in my example
> $f).
>
On Oct 20, 2006, at 7:43 AM, textmate-request(a)lists.macromates.com
wrote:
> Send textmate mailing list submissions to
> textmate(a)lists.macromates.com
>
> To subscribe or unsubscribe via the World Wide Web, visit
> http://lists.macromates.com/mailman/listinfo/textmate
> or, via email, send a message with subject or body 'help' to
> textmate-request(a)lists.macromates.com
>
> You can reach the person managing the list at
> textmate-owner(a)lists.macromates.com
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of textmate digest..."
> Today's Topics:
>
> 1. Re: Executing command on all files in a project AND using TM
> variables in a search and replace (Allan Odgaard)
> 2. Re: Latex: completion with esc not working (Christopher
> Brewster)
> 3. Re: TODO problem (Jasper van der Meulen)
>
> From: Allan Odgaard <throw-away-1(a)macromates.com>
> Date: October 20, 2006 1:42:15 AM EDT
> To: TextMate users <textmate(a)lists.macromates.com>
> Subject: Re: [TxMt] Executing command on all files in a project AND
> using TM variables in a search and replace
> Reply-To: TextMate users <textmate(a)lists.macromates.com>
>
>
> On 19. Oct 2006, at 22:12, J Fishwick wrote:
>
>> Command:
>> eval arr=("$TM_SELECTED_FILES")
>> for (( i = 0; i < ${#arr[@]}; i++ )); do
>> perl -pe 'while ($string =~ m/mc\d+(.*\n.*True)/g) {$string =~
>> s/mc\d+/tf$TM_FILENAME/g;}';
>> done
>>
>> Input: Entire Doc
>> Output: Replace selected text
>
> You need to set input/output to none/discard (or show tool tip, or
> maybe show as HTML and make your command output some progress info).
>
> Then your actual command needs to load the file itself, it already
> iterators over the selected files, it just doesn’t tell perl to
> load/overwrite each of them.
>
> So a simple example would instead be:
>
> eval arr=("$TM_SELECTED_FILES")
> for f in "${arr[@]}"; do
> perl -i.bak -pe 's/bar/bar/' "$f"
> done
> rescan_project # do this after you make changes to files, so TM
> will rescan
>
> This will replace foo with bar in all selected files (and create
> a .bak file with the old contents).
>
> Two other things: 1) inside single quoted strings, shell variables
> are not expanded, so if you want to use $TM_… then you need to use
> double quotes, and 2) TM_FILENAME is the filename of the file open
> when you called the file, it won’t update inside the loop. Here you
> would instead use the current file given by the loop (in my example
> $f).
>
>
>
>
>
>
>
> From: Christopher Brewster <C.Brewster(a)dcs.shef.ac.uk>
> Date: October 20, 2006 6:35:06 AM EDT
> To: TextMate users <textmate(a)lists.macromates.com>
> Subject: Re: [TxMt] Latex: completion with esc not working
> Reply-To: TextMate users <textmate(a)lists.macromates.com>
>
>
>
>> On 19 Oct 2006, at 23:13, Charilaos Skiadas wrote:
>>
>>> Is this what they should be?
>>>
>> That looks right. Is this a saved file? Try it with both saved
>> files as well as untitled documents.
>
> The problem is specific to that project I am afraid. There is no
> problem with other latex files in other folders.
>
>> Is it part of a project, via TM_LATEX_MASTER?
>
> yes it is
>> Are you setting TM_LATEX_BIB? (You don't have to, just checking
>> your settings)
>
> Checking this I realised for Bibtex the path had changed. So the
> \cite{} case is solved.
> But the \ref{} case is still not working (in this project).
>
>> How are you testing that it is not working?
>
> for \ref{}
> put cursor in braces and press esc
>
>> Does the other completion command work, the one via opt-esc?
> This does nothing.
>
>> Make sure that you are up to date, I recently fixed a problem
>> someone else had that sounds possibly similar, though his was with
>> the Command Completion command.
> I am up to date.
>
> Thanks,
>
> Christopher
>
>
> ______________________________________________________________________
> For new threads USE THIS: textmate(a)lists.macromates.com
> (threading gets destroyed and the universe will collapse if you don't)
> http://lists.macromates.com/mailman/listinfo/textmate
>
>
>
>
> From: Jasper van der Meulen <jasper(a)logt.nu>
> Date: October 20, 2006 7:43:02 AM EDT
> To: TextMate users <textmate(a)lists.macromates.com>
> Subject: Re: [TxMt] TODO problem
> Reply-To: TextMate users <textmate(a)lists.macromates.com>
>
>
> I don't think you understand what i mean...
>
> I've got it in a picture now ..
>
> <screenshot 2.png>
> however
> Regards J.
>
>
> On 20-okt-2006, at 7:26, Allan Odgaard wrote:
>
>> On 20. Oct 2006, at 00:02, Jasper van der Meulen wrote:
>>
>>> Could it be that a project is different for the TODO ?
>>
>> It visits all files found under TM_PROJECT_DIRECTORY (using the
>> globally configured folder reference patterns for what to skip).
>>
>>
>>
>>
>> _____________________________________________________________________
>> _
>> For new threads USE THIS: textmate(a)lists.macromates.com
>> (threading gets destroyed and the universe will collapse if you
>> don't)
>> http://lists.macromates.com/mailman/listinfo/textmate
>
>
> _______________________________________________
> textmate mailing list
> textmate(a)lists.macromates.com
> http://lists.macromates.com/mailman/listinfo/textmate
hello:)
In Typeset & View (PDF) i don't understand that :
# Set up TeX compiler, fallback to xelatex if document indicates it
if grep -Esq '\\usepackage{.*(xunicode|fontspec)|program=xelatex' "$M"
then DEF_TEX=xelatex
else DEF_TEX=pdflatex
What is the exact syntax to use XeLaTeX ?
Thanks
Greetings Alain
Hello,
I'm currently using kGTD as my tool of choice, but a pure text-based
solution looks very attracting. I see there are several bundles
mentioning GTD and I was wondering what were their differences and if
they were all maintained.
Additionally, is someone using one of these bundles and somehow
synchronizing their GTD setup to a Palm? If so, I'd be delighted to
hear about it.
Thanks a lot,
Alan
--
Alan Schmitt <http://alan.petitepomme.net/>
The hacker: someone who figured things out and made something cool
happen.
.O.
..O
OOO
Hello,
I'm trying to get the MySQL bundle working so I can browse tables.
I get this on my MacBook Pro:
Traceback (most recent call last): File "/Applications/Document/
TextMate.app/Contents/SharedSupport/Bundles/SQL.tmbundle/Support/bin/
tableBrowser.py", line 284, in ? sys.exit(main()) File "/Applications/
Document/TextMate.app/Contents/SharedSupport/Bundles/SQL.tmbundle/
Support/bin/tableBrowser.py", line 109, in main listTables
(dbName,dbHost,dbPort,serverType,passwd,dbUser) File "/Applications/
Document/TextMate.app/Contents/SharedSupport/Bundles/SQL.tmbundle/
Support/bin/tableBrowser.py", line 130, in listTables mycon =
MySQLdb.connect(db=dbName,host=dbHost,port=int
(dbPort),user=dbUser,passwd=passwd) NameError: global name 'MySQLdb'
is not defined
I tried using tablebrowser.py standalone and it failed due to no
MySQLdb module for Python.
Downloaded and installed MySQLdb and now Python works properly with
the module and MySQL.
Alas, TextMate still can't find MySQLdb and fails with the same error.
I note that the first line of tablebrowser.py is:
#!/usr/bin/env python
My python is at:
/usr/bin/python
Is TextMate using some other Python than my native one?
If not, why can I interactively use Python with MySQL, but TextMate
won't work.
I have all the TM environment variables set so I don't think that is it.
Thoughts?
Thanks,
ml
In all the files in my project, where a certain string of characters is present,
I'm trying to replace a string with the a part of the file name containing said
string.
For example, files are called ace[1-12].xml, and for everytime there is a the
block: "mc\d+(.*\n.*True)" I want to replace mc\d+ with tf0(number prefix from
filename). I know this hits on a couple more adavanced concepts I don't have a
full grasp on yet, but this editor is goading me on to learn regex and basic
command line skills...
The command I have now is:
Save: nothing
Command:
eval arr=("$TM_SELECTED_FILES")
for (( i = 0; i < ${#arr[@]}; i++ )); do
perl -pe 'while ($string =~ m/mc\d+(.*\n.*True)/g) {$string =~
s/mc\d+/tf$TM_FILENAME/g;}';
done
Input: Entire Doc
Output: Replace selected text
It's choking big time... any pointers to the numerous places I'm screwing up?
Thanks~
Hi,
I just built up a Command as BASH which calls a perl script via
TM_BUNDLE_SUPPORT.
The perl script is calling CocoaDialog via $ENV{'TM_SUPPORT_PATH'} and
so forth.
Everything works fine.
Then I installed the same tmBundle at an other Mac. Same OS and TM version.
The funny thing now is that the perl script cannot call CocoaDialog
via TM_SUPPORT_PATH because TM_SUPPORT_PATH is set to
/Users/Bibiko/Library/Application Support/TextMate/Support.
I promise that I didn't set this variable to that path.
On the other hand I tried the bundle command 'Find In all Document'
which is a ruby script. Here it works with the variable
TM_SUPPORT_PATH. !!??!!
The only difference is that this ruby script is not written as a
separate file.
Please, help is needed.
If I use the command 'Show all TM.*' the TM_SUPPORT_PATH is also set
to this weird path.
Thanks in advance
Hans
----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
Is there a way to close all open tabs in a project without closing the
project? Is there some way I could create this? As a macro maybe? I'd
like to be able to map this to something like cmd+shift+W.
-dave
In all the files in my project, where a certain string of characters
is present,
I'm trying to replace a string with the a part of the file name
containing said
string.
For example, files are called ace[1-12].xml, and for everytime there
is a the
block: "mc\d+(.*\n.*True)" I want to replace mc\d+ with tf0(number
prefix from
filename). I know this hits on a couple more adavanced concepts I
don't have a
full grasp on yet, but this editor is goading me on to learn regex
and basic
command line skills...
The command I have now is:
Save: nothing
Command:
eval arr=("$TM_SELECTED_FILES")
for (( i = 0; i < ${#arr[ <at> ]}; i++ )); do
perl -pe 'while ($string =~ m/mc\d+(.*\n.*True)/g) {$string =~
s/mc\d+/tf$TM_FILENAME/g;}';
done
Input: Entire Doc
Output: Replace selected text
It's choking big time... any pointers to the numerous places I'm
screwing up?
Thanks~
hello
if i create a new file with Latex bundle and if i drag a .pdf file,
the result is correct
\begin{figure}[htbp]
\begin{center}
\includegraphics[height=3in]{/Users/ego/Desktop/
AmeriqueNordES2005.pdf}
\end{center}
\caption{caption}
\label{fig:label}
\end{figure}
but if i save the .tex file and i want to drag again a .pdf file :
the result is /usr/lib/ruby/1.8/pathname.rb:529:in ??? strange
Thanks for a solution
Alain Matthes
TextMate Version 1.5.3 (1292)
Hi,
Sorry for this basic question, but I'm a textmate newbie and I just
couldn't find a way to do this. I was editing a HTML file and then I
accidentally pressed some key combination and now the syntax
highlighting is messed up. I tried finding a way of going back to HTML
highlighting mode to no avail. How do I do this?
Thanks,
Yariv
Hi there,
One thing I miss from my XEmacs days, was the
“etags” command which is very useful when you
have a directory full of Fortran files…
If, in a fortran file, you had
call foobar(blah blah)
It would automatically open the file in which
the function foobar was defined.
So I made a shell script which mimic that.
http://guerom00.free.fr/clutter/Fortran_Tags.sh
Just create a new command which reads :
Wherever_you_put_the_script/Fortran_Tags.sh "$TM_DIRECTORY"\\
"$TM_CURRENT_WORD" &>/dev/null &
It will create a tags file named “.tags” in the current
directory.
I know it's rather ugly and not perfect but seems
to work OK… Hopefully one person or two will
find this usefull :-D
Hi there,
I made myself a simple script to compile Fortran
programs using the “make” command. I wanted the
compilation window to automatically close on
successful compilation, so this script basically
reads :
htmlHeader "Compiling current programs using Make"
echo "<h1>Starting Fortran Compile</h1>"
echo
cd "$TM_DIRECTORY"
make|pre
status=$?
echo "status = "${status}
if [[ "${status}" -ne "0" ]]; then
echo "<h1>Compilation Failed</h1>"
else
echo "<h1>Compilation Successful</h1>"
# closeWindow
fi
htmlFooter
The problem is that, even when the compilation
fails, the $status is equal to 0…
http://img220.imageshack.us/img220/8341/image1po1.jpg
What did I do wrong ? How do I fix this ?
Thanks in advance.
PS : I don't know if I'm the only one, but using Safari
for the RSS feed for this mailing list, I receive EVERY
message twice… Rather annoying…
>> [...]
>> make|pre
>> [...]
>> The problem is that, even when the compilation
>> fails, the $status is equal to 0
>>
>
> The problem is that when you do ‘foo|bar’ then the result of the
> entire command is that of the last command in the chain.
You can use the bash PIPESTATUS array.
j.
I've come across this a couple of times now: a file which looks fine
in TextEdit or BBEdit loses its line breaks when opened in TextMate.
Looking at the problem file in HexEdit, the following Hex is being
used for linebreaks: E2 80 A8, which appears to be a UTF-8 linebreak,
based on some preliminary googling. Saving a short UTF-8 file in TM,
the linebreaks are: 0A (\n).
Would it be possible for TM to support this format of linebreaks, at
least on read?
jb
I work with a lot of code which has 'dropped' braces, such as
if ()
{
}
else
{
}
I personally prefer the K&R style, such as
if () {
} else {
}
I made this new macro to make conversions easier - it simply keeps a
space before/after the brace when joining lines - otherwise joining
the lines would result in
if (){
}else{
}
I added it as a second macro with a scope of just "source", so text
files won't be affected.
Thoughts?
Cheers,
Ciarán
Hi,
I had written about this before, but I am still seeing this.
If I have a mounted volume (Either AFP, or in the latest case NFS),
and I've opened some files from that volume (and have subsequently
closed them), when I go to dismount that
volume, I get an error in the Finder stating that the volume is in use.
It always end up being Textmate. Using the terminal, I can see:
john$ lsof | grep TextMate
TextMate 376 john cwd VDIR 26,9 4096 343647
/Volumes/10.1.1.104/public_html
(The volume that is mounted is just /Volumes/10.1.1.104/, public_html
is the directory I happened to open a file from.)
The only way around this is to quit TextMate. Any ideas why TM in
holding onto the volume like that?
Thanks,
jt
Hi Allan,
thanks for your help!
this is a follow up to a newsgroup post from yesterday. I am having
problems with the blogging bundle.
I turned the xmlrpc logging on - see below.
On 16-Oct-06, at 11:58 PM, textmate-request(a)lists.macromates.com wrote:
> On the server, you need to create a writable (by your apache user)
> directory for the uploads and set that in the WordPress preferences.
"the uploads" meaning for example jpg files that I would upload with
a blog post?
I can create a directory on my server (that is not my own apache
server, but a hosted one) and I can probably find the corresponding
setting in wordpress... but I guess the following issue should be
resolved first.
>
> Is this a 100% untampered WordPress 2.0 install?
>
> In your install of WP is a xmlrpc.php file, set $xmlrpc_logging =
> 1; in that file and check what the XML-RPC log says when you post,
> i.e. if TM provides the post in two chunks. If it does, it is your
> WP install which needs fixing.
The xmlrpc.log looks like this. With "two chunks" - do you mean the
part highlighted in red?
2006-10-17 23:05:49 Input: <?xml version="1.0" ?
><methodCall><methodName>metaWeblog.editPost</
methodName><params><param><value><string>50</string></value></
param><param><value><string>Claus</string></value></
param><param><value><string>Butterfly2010</string></value></
param><param><value><struct><member><name>mt_allow_pings</
name><value><string>open</string></value></
member><member><name>mt_allow_comments</name><value><string>open</
string></value></member><member><name>title</
name><value><string>Blogging via TM</string></value></
member><member><name>categories</
name><value><array><data><value><string>water in the media</string></
value></data></array></value></member><member><name>description</
name><value><string>This is a blog entry written from TM
Holladrihia</string></value></member></struct></value></
param><param><value><boolean>1</boolean></value></param></params></
methodCall>
2006-10-17 23:05:49 Output: (MW) Edited ! ID: 50
2006-10-17 23:05:50 Input: <?xml version="1.0" ?
><methodCall><methodName>metaWeblog.getPost</
methodName><params><param><value><string>50</string></value></
param><param><value><string>Claus</string></value></
param><param><value><string>Butterfly2010</string></value></param></
params></methodCall>
2006-10-17 23:06:38 Input: <?xml version="1.0" ?
><methodCall><methodName>metaWeblog.editPost</
methodName><params><param><value><string>50</string></value></
param><param><value><string>Claus</string></value></
param><param><value><string>Butterfly2010</string></value></
param><param><value><struct><member><name>mt_allow_pings</
name><value><string>open</string></value></
member><member><name>mt_allow_comments</name><value><string>open</
string></value></member><member><name>title</
name><value><string>Blogging via TM</string></value></
member><member><name>categories</
name><value><array><data><value><string>water in the media</string></
value></data></array></value></member><member><name>description</
name><value><string>This is a blog entry written from TM
Holladrihia
</string></value></member><member><name>mt_text_more</
name><value><string>
This now should be below the cut! It better damn be!</string></
value></member></struct></value></param><param><value><boolean>1</
boolean></value></param></params></methodCall>
2006-10-17 23:06:38 Output: (MW) Edited ! ID: 50
Thanks again,
Claus
Hello all,
I am just installing TextMate on my new iMac and I face la little problem : by default, spell checking is activated. My question : how can I disable spell checking ?
Thanks
G
___________________________________________________________________________
Découvrez une nouvelle façon d'obtenir des réponses à toutes vos questions !
Demandez à ceux qui savent sur Yahoo! Questions/Réponses
http://fr.answers.yahoo.com