On 11/08/2015 03:19 PM, Lex Trotman wrote:
On 9 November 2015 at 05:34, Devyn Collier Johnson
<devyncjohnson@gmail.com> wrote:

Geany Dev Team:

I have some additional ideas that I would like to share with you all and get
feedback before I implement the code.

./data/filetype_extensions.conf

- Adding "*.s03;*.s79;*.s82;*.s90;*.s;*.S;" to "ASM="
*.s## - GNU-style Assembly (GAS/AT&T)
Capital "S" - Must be preprocessed
Lowercase "s" - Does not require preprocessing
The numbers appended to *.s indicate the assembly version or intended
microprocessor
*.s is also used by Common Intermediate Language (CIL) which looks like
assembly
The Linux kernel contains some "*.S" files such as
https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git/tree/arch/x86/boot/header.S?id=refs/tags/v4.3
Sources:
http://labor-liber.org/en/gnu-linux/development/extensions
http://wiki.linuxquestions.org/wiki/List_of_file_extensions#S
https://en.wikipedia.org/wiki/Common_Intermediate_Language

The geany ASM parser specifies .s and .S so those should be ok, if
they don't work for some special cases (eg CIL) that can't be helped.
As for the other ASMs, well, if they work why no.  Note that both the
lexer and parser are pretty general, so there will be no
specialisation for different extensions.



- Adding "*.ll;" to "ASM="
I suggested this before, but I want to ensure that the team is okay with
this idea before I commit and PR
LLVM assembly ( http://linux.die.net/man/1/llvm-as )

Well, if it works, though I am not sure if many people will edit LLVM.
Remember that if we provide something by default we are advertising
that it should work, and that users can reasonably raise bug reports
if it doesn't.  But we won't be able to support them.  For obscure
things I would prefer that the few who need it can add it to their own
`filetype_extensions.conf`.



- Adding "*.i;" to "C="
*.i - C source code which should not be preprocessed
http://labor-liber.org/en/gnu-linux/development/extensions

g++ treats this as C++, all languages are *not* C :)

This is the same problem as .h, it could be either.


- Adding "*.ii;" to "C++="
*.ii - C++ source code which should not be preprocessed
http://labor-liber.org/en/gnu-linux/development/extensions

- Adding "*.r;*.f15;*.F15;" to "Fortran="
*.r - Fortran source code which must be preprocessed with a RATFOR
preprocessor
http://labor-liber.org/en/gnu-linux/development/extensions
*.f15 - Fortran 2015
http://fortranwiki.org/fortran/show/File+extensions

I very much doubt that lexer and parser support these variants of
fortran properly.


- Adding "*.xaml;" to "XML="
XAML (Extensible Application Markup Language)
This is a markup-language made by Microsoft
XAML is used in .NET Framework 3 and 4
Mimetype = application/xaml+xml
https://en.wikipedia.org/wiki/Extensible_Application_Markup_Language
https://msdn.microsoft.com/en-us/library/cc295302.aspx
https://msdn.microsoft.com/en-us/library/System.Windows.Markup.aspx

Myeh, we should remove XML totally </reaction="allergic to XML"> :)


Previous Ideas

These proposed Python file-extensions are not officially mentioned in the
Python Documentation. I shared this idea previously in the mailing list, but
I would like to know if this idea is a possibility or a definite "no".
However, some programmers use the extensions.
- py2 and py3
Used to specifically distinguish Python3 code from Python2, especially when
code is written that only works on one of the two versions.
- pygtk
Used to specifically show that a script contains PyGObject or PyGTK code, as
opposed to Qt
- pyqt
Like pygtk, but used to specifically show that a script contains PySide or
PyQt code

After reading my explanation, what do you think about the suggested Python
extensions?

Well, IMHO these should be rare, since they are not supported by the
standard Python implementation, so again the users can add them to
their personal `filetypes.extensions.conf` if they really want them.



Possible Ideas

These are ideas that I am thinking about (or would like to hear your
thoughts) before I proceed.

- Previously, I think mentioned adding XNA support. That idea will be
delayed or never implemented because Microsoft discontinued XNA. In
addition, I cannot find Microsoft's XNA specification. However, I would be
willing to add XNA support if the Geany Dev Team thinks it would still be a
good idea.


Unless you or someone else is clamouring for support of this
(discontinued you say) language then I don't see the point, again it
has to be supported.

This point applies to all the following as well.  Things should not be
added "just because they exist".  There needs to be a demand for it to
be available in Geany, it needs to be sensible, and some indication
that the added features will be supported.

Rather too much gets dumped into open source projects and then left to
fester </rant> :)

Many of the following are things that don't get edited often, or have
better tools elsewhere, eg qtcreator for qml.


- I would like to add support for Qt's QML (Qt Meta Language or Qt Modeling
Language). It is a scripting language that resembles JavaScript (
http://doc.qt.io/qt-5/qmlapplications.html ). If the Geany lexers and
parsers highlight this language well, would it be a good idea to add it?

- Would it be a good idea to add Java bytecode files (*.class) to "ASM="? I
do not know Java, so this may be a poor idea.
https://en.wikipedia.org/wiki/Java_bytecode#Example

- In my opinion, it would be a good idea to add support for sed scripts (
http://www.grymoire.com/Unix/Sed.html#uh-20 &&
https://www.gnu.org/software/sed/manual/sed.html#Centering-lines ). They use
the "*.sed" file-extension. However, I will need to experiment with the
lexers and parsers.

- In addition, I could add awk scripts (
http://www.grymoire.com/Unix/Scripts/awk_print_squares.awk ). They use the
"*.awk" file-extension. However, I will need to experiment with the lexers
and parsers.

- Add support for FASTA ( https://en.wikipedia.org/wiki/FASTA_format &&
https://en.wikipedia.org/wiki/FASTA &&
http://fasta.bioch.virginia.edu/fasta_www2/fasta_list2.shtml ). I am not
sure how you will all feel about this idea. FASTA is a file format used to
represent nucleotide and peptide sequences. True, it is not a programming
language, and I do not know yet if any lexer or parser used by Geany will
work with FASTA. However, with the rise of "open-source biology", Boolean
Integrase Logic (BIL) gates, and biochemical-based transistors (DNA and RNA
transistors = transcriptors), this may be an idea to consider. Also, such a
feature would make Geany stand-out among other "traditional" IDEs.


Geany stands out by being fast and lightweight, adding large amounts
of rarely used excess baggage will kill that.

Cheers
Lex


--
Thanks,
Devyn Collier Johnson
DevynCJohnson@Gmail.com


_______________________________________________
Devel mailing list
Devel@lists.geany.org
https://lists.geany.org/cgi-bin/mailman/listinfo/devel


_______________________________________________
Devel mailing list
Devel@lists.geany.org
https://lists.geany.org/cgi-bin/mailman/listinfo/devel
Geany Dev Team:

Thank you Lex Trotman and Matthew Brush for your time, effort, and insightful feedback.

Thanks Matthew for the QML link; I greatly appreciate it. I added the code to my system. Could we add this to the mainstream Geany? True, QtCreator may have better support, but some people prefer Geany over all IDEs (including me).

Okay. I just thought I could at least suggest FASTA. To me, it seems like a good idea, although very odd and unusual for a programming IDE. Sometimes, I like to make wild ideas because such ideas may become a "brilliant idea". However, I understand the team's viewpoint.

True, one of  Geany's best features is that it is fast and light-weight. However, are there any benchmarks that indicate a significant performance drop when adding many file-extensions and file-types?

Perhaps, a "plugin" could be made that has support for extra and "weird" file-extensions and file-types. Could Geany be modified to also look for a "filetypes_extension_extra.conf" file that would contain such file-extensions? For instance, an additional package could be made that contains "filetypes_extension_extra.conf" and various filetypes.*.conf files. If the user downloads and installs this package, Geany will see "filetypes_extension_extra.conf" and merge the contents to "filetypes_extension.conf" and load the related filetypes.*.conf files. Such a feature may help us with development. For instance, we could use such a feature for testing proposed file and language support and provide "backported" and unstable file-extension support. As a specific example, the Manpage file-extensions I proposed could be added to such a package until Scintilla adds highlighting support for Manpages. Do you understand what I am saying? Would this be a possible idea to consider?

Will Not Add

These are ideas that I will not add after hearing the team's viewpoints and after I have reconsidered and pros and cons.

- *.ll: Good point, Matthew and Lex.
- Python: Okay, I will not worry about the suggested extensions.
- *.r: Thank you, Matthew, for pointing that out. I knew that R used the "*.R" extension, but I was unaware of "*.r".
- XNA: I do not know XNA very well neither do I personally care about XNA. I just wanted to ensure that Geany supports XNA for those that do like XNA. However, since XNA is discontinued and I cannot find the specification, I will not add XNA. Unless, in the future, support is desirable.
- Java Bytecode - I did more research, and I feel that this idea can be dropped.

Lex, I assume you are joking about removing "XML=", right? In my opinion, XML support in Geany is a "must".

Will Add Soon

These are ideas that I feel are beneficial and that the team approves.

- Add "*.f15;*.F15;" to "Fortran="
*.f15 = Fortran 2015
I do not know Fortran, but if highlighting does not work well with Fortran 2015, perhaps it will when Scintilla adds updates for this new Fortran standard
http://fortranwiki.org/fortran/show/File+extensions

- Add "*.xaml;" to "XML="
XAML (Extensible Application Markup Language)
This is a markup-language made by Microsoft
XAML is used in .NET Framework 3 and 4
Mimetype = application/xaml+xml
https://en.wikipedia.org/wiki/Extensible_Application_Markup_Language
https://msdn.microsoft.com/en-us/library/cc295302.aspx
https://msdn.microsoft.com/en-us/library/System.Windows.Markup.aspx

- Adding "*.ii;" to "C++="
*.ii - C++ source code which should not be preprocessed
http://labor-liber.org/en/gnu-linux/development/extensions

May Add/Consider

These are ideas that I may reconsider, or I would like to hear your further thoughts.

- *.s: True, the ASM lexer does not support AT&T syntax. However, if Geany recognizes the file-type, then the Assembly tools listed in the "Build" menu are available.
- *.i: *.i and *.ii are different. "i" is for C, and "ii" is for C++
- *.sed and *.awk: Sed and Awk are commonly used on Unixoid (Unix systems and Unix-like systems) systems. Are you sure that this idea should be dropped?

Additional Comments

By the way, I would like to thank everyone in the Geany Dev Team for their time, effort, and support. I am also thankful for being a part of the team.

What must I do to earn a spot under "Contributors" in Geany's "Credits" tab of the "About" window?

--
Thanks,
Devyn Collier Johnson
DevynCJohnson@Gmail.com