source_codepage autodetect with enca program.

[ASM]

Hello.

I wrote stupid patch for autodetect source_codepage. It working then open/view/edit events.

It using enca package (​http://gitorious.org/enca, for fedora: ​https://admin.fedoraproject.org/pkgdb/packages/name/enca)

Use autodetect source_codepage very convenient. Please don't ignore it.

[ASM]

Fix, if codepage don't detect, don't warning.

[osgx]

There must be way to disable auto-detection or to manually select codepage if enca fails to do it right.

[slavazanko]

Yes, right. But this is an enhancement (not a bugfixing or code cleanup). Therefore this ticket not for '4.7.0' milestone. Just await :)

[ASM]

I tested my patch and made a few notes:

• Need add on MC menu:

Enca on/off
Enca language
Maybe add default source_codepage encoding (if enca fail)

• Don't use mc.ext for enca. Is unnecessary. Program 'file' is often mistaken for utf-8 files, view their 'data'. If enca is on, need exec it for all source files. If enca wrong, need set display_codepage(or another).

Another idea?

There must be way to disable auto-detection or to manually select codepage if enca fails to do it right.

Is't work. (F9->Command->Encoding)

---
I'm no good code writer, but I'm working on that. I try rewrite patch.
Sorry my english, but I'm working on that too. :-)

[slavazanko]

ASM: is you have own publish git-repo? For example, on ​http://github.com/ or on
​http://repo.or.cz/

This better to develop in your own branch, because in this case you will maintainer of your idea and anybody will send patches to you instread of... :)

[ASM]

Hello, folks!

I add enca support (autodetect). Need add to ini file new option.
What name give this? How to use?

For example something like:

• codepage_autodetect= no use enca
• codepage_autodetect=ru use enca, set language Russian
• codapage_autodetect=off no use enca

[ASM]

codepage_autodetect -> autodetect_codeset

[ASM]

I release it in ​http://github.com/BASM/mc-basm/tree/ASM/1838_enca.

• added autodetect parameter,
• added enca support,
• added default codepage, if enca fail.

[slavazanko]

Please, use one-style source coding

[ASM]

• Style code fixed,
• Bugs fixed.

I think autodetect language done.

[ASM]

s/language/encoding/

[ASM]

Fixed version.

[ASM]

My branch released in: ​http://github.com/BASM/mc-basm/tree/1838_enca

[slavazanko]

Created & rebased branch 1838_add_enca_support

Initial changeset:b5b119ffac2bca5edb30aaec98a32b6fa9bab031

Review, please.

[angel_il]

Changeset: 71d58780d127cc8d2426e00b40f8ffc0a4854c9d (forced update)

[andrew_b]

Code was fixed and documentaion was updated. Please revire again.
changeset:dd93a53afc6ec190a95ef6d83b9a72e9cea5f3e4 -- code
changeset:cc432bd67f4c2edfecd39b60e22f6158d5127005 -- documentation

[slavazanko]

merged into master: ac60804870b9af77a6f9e48e72c85e9efd182e90

[andrew_b]

Charset autodetection is partially broken in recent master (4.7.2-46-g7843203).
In my KOI8-R locale 8-bit the UTF-8 charset is not autodetected as in editor as in viewer. 8-bit locales are autodetected.

To fix this issue, the 1838_codeset_autodetect_fix branch was created. Parent branch is master.
Initial changeset:e053e1d29d79d92de96d0fcdaae1cfd231d6bcd9

[andrew_b]

Merged to master.
changeset:12969b49741caed5bf529e062cc47d8edb8802ad

git log --pretty=oneline 6117c5b..12969b4