Using 1.5.7 (1464), I realized that the command-shift-A shortcut for Add
New Files... doesn't work to add files to a project. I just get a
system beep.
I've seen no discussion of this here, aside from a 2005 post that said
it should start working once you've manually displayed the menu, so I'm
guessing this isn't a common problem. Yet I'm seeing it on two
different Macs. Command-Shift-N, which is on the File menu as well as
the gear menu, does work. Command-shift-G, which is also on the gear
menu (and not any other menu), doesn't work either.
Is this in fact a known problem? Is there a workaround, other than
QuicKeys/etc? I took a look at KeyBindings.dict, but I'm not sure if I
can add command-keystrokes that way, or what the setting would be
called, etc.
Jay Levitt
I've recently begun learning Erlang, and in the process made some
updates to the Erlang bundle that might be useful to others. I've
created an AppleScript that can perform basic remote control of a
Terminal window running an Erlang shell -- enough to compile a source
file at least.
There's a short screencast and download information on my blog:
http://imbriaco.com/braindump/articles/first-post-erlang-and-textmate/
If the original author of the Erlang bundle reads this, I'd love to
contribute this back to the official bundle. Feel free to contact me
off list.
-Mark
I have asked about this in the obvious places (pgf help forum and
comp.text.tex) and after some initial help, the help has dried up. As
this is going to be a specifically Mac users here, maybe the problem
may be more obvious.
When I try and compile this code (from the manual):
\documentclass[a4paper]{article}
\usepackage{pgfplots}
\begin{document}
\begin{tikzpicture}%
\begin{axis}[
xlabel=$x$,
ylabel=$\sin(x)$,
name=an axis,
]
\addplot plot [id=sin] function{sin(x)};
\end{axis}
\end{tikzpicture}%
\end{document}
I get these errors:
Latex Error: ./PDFPlotTrial2.tex:11 Package pgfplots Error: Sorry, the
gnuplot-result file 'PDFPlotTrial2.sin.table' could not be found.
Maybe you need to enable the shell-escape feature? For pdflatex, this
is '>> pdflatex -shell-escape'. You can also invoke '>> gnuplot
PDFPlotTrial2.sin.table' manually..
! Package pgfplots Warning: You have a plot with empty range.
Replacing it with default and clearing plots.
Latex Error: ./PDFPlotTrial2.tex:13 Extra }, or forgotten \endgroup.
Latex Error: ./PDFPlotTrial2.tex:13 Missing } inserted.
Latex Error: ./PDFPlotTrial2.tex:13 Extra }, or forgotten \endgroup.
Latex Error: ./PDFPlotTrial2.tex:13 Missing } inserted.
Latex Error: ./PDFPlotTrial2.tex:15 LaTeX Error: \begin{tikzpicture}
on input line 5 ended by \end{document}.
I have run pdflatex --shell-escape PDFPlotTrial2.tex from terminal
(which seemed to work as I got pages of text appearing with no obvious
error messages) and I have upgraded to version 2 of pgf. GNUPlot was
installed with fink and runs OK from typing gnuplot in terminal.
I am now stuck, and no idea what to try next. I am pretty new to this
so it could well be something so obvious that no one is suggesting it.
Can anyone help.
Many thanks,
Graham
Hi all-
I'm working on a language grammar for Forth, and am hoping I could get
some advice or pointers for my work. I've read the TextMate manual, and
James Gray's fine book, although I'm sure I haven't extracted every
nugget of info from those texts.
As some of you may be aware, Forth is a stateful language, and there
are certain keywords that are valid only when Forth is compiling. For
example, some words, like IF, THEN and ELSE, can be used only within a
word (function) definition.
So it would be nice to capture this in the scope, and assign these
compile-only words to an "invalid" scope when they are used outside a
function definition. The question is, how to (best) do this?
I've looked at the C language definition a bit, and clearly there are
two modes there: "preprocessing", and "not". But with Forth, I should
be able to assign an invalid scope to the top-level; in essence, have
an invalid scope for source.forth.
Any thoughts?
Best, Charles
Hi,
I've just upgraded the standard Java bundle and TM's support lib for
executor stuff.
I have the following tiny java script:
import java.io.*;
public class test {
public static void main(String args[]){
System.out.println("Hello World");
}
}
saved as test.java
If I compile and run it on a console, no problem.
If I use the Java bundle's "Compile & Run" I get this:
/Applications/TextMate.app/Contents/SharedSupport/Support/lib/tm/
executor.rb:87:in `run': undefined method `call' for nil:NilClass
(NoMethodError) from /Applications/TextMate.app/Contents/SharedSupport/
Support/lib/io.rb:41:in `call' from /Applications/TextMate.app/
Contents/SharedSupport/Support/lib/io.rb:41:in `exhaust' from /
Applications/TextMate.app/Contents/SharedSupport/Support/lib/io.rb:
41:in `each_line' from /Applications/TextMate.app/Contents/
SharedSupport/Support/lib/io.rb:41:in `exhaust' from /Applications/
TextMate.app/Contents/SharedSupport/Support/lib/tm/process.rb:165:in
`run' from /Applications/TextMate.app/Contents/SharedSupport/Support/
lib/tm/executor.rb:99:in `run' from /Applications/TextMate.app/
Contents/SharedSupport/Support/lib/tm/executor.rb:198:in
`process_output_wrapper' from /Applications/TextMate.app/Contents/
SharedSupport/Support/lib/tm/executor.rb:98:in `run' from /
Applications/TextMate.app/Contents/SharedSupport/Support/lib/tm/
htmloutput.rb:134:in `call' from /Applications/TextMate.app/Contents/
SharedSupport/Support/lib/tm/htmloutput.rb:134:in `show' from /
Applications/TextMate.app/Contents/SharedSupport/Support/lib/tm/
executor.rb:83:in `run' from /Applications/TextMate.app/Contents/
SharedSupport/Bundles/Java.tmbundle/Support/bin/javamate.rb:32
I didn't set up any env variables.
I'm running a Intel MacOSX 10.5.4 (very fast;) with TM Version 1.5.7
(1466)
With the 'old' Java bundle ("Compile & Run single script") it works.
Any help available?
The new bundle has a nice feature. "Compile & Run (with args)"
I couldn't try out it yet but if I have a folder containing my Java
script and aFile, and I want to do something with that aFile. Is it
possible to enter only the filename aFile as argument without the
absolute path. In other words, does run the script in that folder?
Many thanks in advance,
--Hans
This Python code:
#!/usr/bin/env python
# encoding: utf-8
import sys
import os
import locale
def main():
print locale.setlocale(locale.LC_ALL, '')
print locale.format("%d", 282929992, grouping=True)
if __name__ == '__main__':
main()
produces the expected result when run under Python 2.5.2 from bash
shell on Mac OS X:
en_US.UTF-8
282,929,992
and this unexpected result when run from PyMate:
PyMate r8111 running Python 2.5.2 (/usr/bin/env python) >>> loc.py
C/en_US.UTF-8/C/C/C/C
282929992
Notice that the integer is formatted into comma-separated groups when
run from the shell, but not when run from PyMate.
Changing this line:
print locale.setlocale(locale.LC_ALL, '')
to this:
print locale.setlocale(locale.LC_ALL, 'en_US.UTF-8')
gives the expected results in PyMate.
Dave
Hi all-
I'm trying to match character constants of the form:
&a &B &!
'c' 'D' '!'
"e" "F" "@"
^G ^h ^+
within the match field of a language grammar.
If I use the search dialog box, this regexp will work for characters
enclosed in single apostrophes:
((?<=^)|(?<=\s))(([&^]\S)|(\'\S\'))(?=\s)
But it doesn't work when I get it into the match field, I suspect
because of the framing apostrophes, as in:
{ name = 'constant.character.forth';
match = '((?<=^)|(?<=\s))(([&^]\S)|(\'\S\'))(?=\s)';
},
Any ideas about what I'm doing wrong, or how I can solve my trouble?
I should also point out that the ungainly:
((?<=^)|(?<=\s))
seems to be necessary inside a language grammar as (?<=\s) alone
doesn't work, although it does in the search dialog. It matches within
a line, but not at the beginning of a line. It may be that this is only
the case with begin/end matching, and not single lines...
Thanks! Charles
The syntax highlighting for LaTeX math mode is very good. I have
noticed, however, a discrepancy in the way the highlighting and the
automatic delimiters are treated. I mean, normally typing a $ will
automatically generate a second one (after the cursor) to end the math
mode. However, I have found that this fails when a command in math
mode spans multiple lines. That is, if there is a hard return in math
mode, then the delimiters behave incorrectly eventhough the syntax
highlighting is correct.
Here's an example:
$math stuff
that spans two lines$ when I put a $ here it thinks I'm closing the
math mode from the first dollar sign on this line of code rather than
thinking I'm starting a new math mode.
Keith Penrod
Hi there,
Some people are working on updating Python and HTML-related bundles
for Django, prior to Django's 1.0 release (http://www.bitbucket.org/bkerr/django-textmate-bundles/
), and I've got a bundle command problem where I'm hoping someone can
point the way.
The idea is to provide some kind of automation for importing commonly-
used functions and classes from Django's source code. Many of these
objects are hidden fairly deep inside one module or another, and it's
easy to forget which, leading to long tromps through the source code.
I'd like to provide a command, perhaps similar to the HTML bundle's
code completion command, which lets people either a) hit command-
escape after the name of an object and view/paste its import path or
b) start typing 'from django.', hit a key combo, and see a list of
child modules, ideally letting them drill down into the modules until
they find the object they're looking for.
I haven't written complicated bundle commands before and am not sure
how to start this, in particular whether to go with route a or b. I
can either manually build up a dictionary of objects and their import
paths and then maintain that dict as Django evolves, or I can try for
a command that actually searches Django's source code for the
definition of an object. I'm leaning towards the former for two
reasons: Having a list of objects would also serve for syntax
highlighting, and also many objects are defined in one module, then
imported into the __init__ file of another. They're 'supposed' to be
imported from that second module, but a search function wouldn't know
that.
Anyway, any general pointers and issues to consider would be much
appreciated. I'd be writing the script in Python, naturally!
Thanks,
Eric
Using the latest Java bundle, strings are not highlighted correctly:
public class Foo
{
void bar(Object baz)
{
throw new RuntimeException(baz.toString() + "; void");
}
}
Trevor
