On Mon, Jan 21 2019, Edd Barrett <[email protected]> wrote:
> Hi,
>
> On Mon, Jan 21, 2019 at 07:22:40AM +0100, Antoine Jacoutot wrote:
>> Fails to patch.
>
> Indeed!
>
> It looks like this was caused by a bogus Unicode translations at `cvs
> commit` time. I committed this at the same time as TeX Live, which I
> always commit directly from my $HOME on the cvs server (such large
> commits suck over the network). I don't have LC_CTYPE set there; I think
> this was the cause.

Huh, if that is the case, it sucks...

Note that your diff below doesn't apply as-is for me. Since it contains
non-ascii chars you may prefer to send it as an attachment.

[...]
Hmm...  The next patch looks like a unified diff to me...
The text leading up to this was:
--------------------------
|Index: patches/patch-src_synctex_parser_c
|===================================================================
|RCS file: /cvs/ports/editors/texworks/patches/patch-src_synctex_parser_c,v
|retrieving revision 1.1
|diff -u -p -r1.1 patch-src_synctex_parser_c
|--- patches/patch-src_synctex_parser_c 20 Jan 2019 11:20:45 -0000      1.1
|+++ patches/patch-src_synctex_parser_c 21 Jan 2019 09:36:26 -0000
--------------------------
Patching file patches/patch-src_synctex_parser_c using Plan A...
Reversed (or previously applied) patch detected!  Assume -R? [y] 

> Here's a fix, which I will commit from my workstation with the locale
> set correctly. While I was here, also noticed we need a stronger TeX
> Live dependency.
>
> I think no bump is necessary, because it hasn't packaged since the last
> bump. Right?

Yup.

> OK?

OK with whatever mojibake you need to get a working package, but please
test a build with ports-gcc. ;)

> Index: Makefile
> ===================================================================
> RCS file: /cvs/ports/editors/texworks/Makefile,v
> retrieving revision 1.44
> diff -u -p -r1.44 Makefile
> --- Makefile  20 Jan 2019 11:20:42 -0000      1.44
> +++ Makefile  21 Jan 2019 09:38:33 -0000
> @@ -50,7 +50,7 @@ LIB_DEPENDS =               ${MODQT5_LIB_DEPENDS} \
>                       x11/qt5/qtscript
>  LIB_DEPENDS-main =   ${LIB_DEPENDS} \
>                       textproc/hunspell \
> -                     print/texlive/base \
> +                     print/texlive/base>=2018 \
>                       print/poppler,,-qt5
>  LIB_DEPENDS-lua =    ${LIB_DEPENDS} \
>                       ${MODLUA_LIB_DEPENDS}
> Index: patches/patch-src_synctex_parser_c
> ===================================================================
> RCS file: /cvs/ports/editors/texworks/patches/patch-src_synctex_parser_c,v
> retrieving revision 1.1
> diff -u -p -r1.1 patch-src_synctex_parser_c
> --- patches/patch-src_synctex_parser_c        20 Jan 2019 11:20:45 -0000      
> 1.1
> +++ patches/patch-src_synctex_parser_c        21 Jan 2019 09:36:26 -0000
> @@ -120,7 +120,7 @@ Index: src/synctex_parser.c
>  +/**
>  + *  \file synctex_parser.c
>  + *  \brief SyncTeX file parser and controller.
> -+ *  - author: Jérôme LAURENS
> ++ *  - author: Jérôme LAURENS
>  + *  \version 1.19
>  + *  \date Mon Apr 24 07:08:56 UTC 2017
>  + *
> @@ -644,7 +644,7 @@ Index: src/synctex_parser.c
>  - *  It takes care of sending the proper message if possible.
>  +/*  We ensure that SYNCTEX_BUFFER_SIZE < UINT_MAX, I don't know if it makes 
> sense... */
>  +/*  Actually, the minimum buffer size is driven by integer and float 
> parsing, including the unit.
> -+ *  ±0.123456789e123??
> ++ *  ±0.123456789e123??
>    */
>  -#   define SYNCTEX_MSG_SEND(NODE,SELECTOR) if (NODE && 
> SYNCTEX_CAN_PERFORM(NODE,SELECTOR)) {\
>  -            (*((((NODE)->class))->SELECTOR))(NODE);\
> @@ -4202,7 +4202,7 @@ Index: src/synctex_parser.c
>  +}
>   
>  -/*  Actually, the minimum buffer size is driven by integer and float 
> parsing.
> -- *  ±0.123456789e123
> +- *  ±0.123456789e123
>  - */
>  -#   define SYNCTEX_BUFFER_MIN_SIZE 16
>  -#   define SYNCTEX_BUFFER_SIZE 32768
> @@ -9422,14 +9422,14 @@ Index: src/synctex_parser.c
>  +
>  +/**
>  + *  The higher box node in the parent hierarchy which
> -+ *  mean line number is the one of node ±1.
> ++ *  mean line number is the one of node ±1.
>  + *  This enclosing box is computed as follows
>  + *  1) get the first hbox in the parent linked list
>  + *  starting at node.
>  + *  If there is none, simply return the parent of node.
>  + *  2) compute the mean line number
>  + *  3) scans up the tree for the higher hbox with
> -+ *  the same mean line number, ±1 eventually
> ++ *  the same mean line number, ±1 eventually
>  +*  - parameter node: a node.
>  + *  - returns: a (proxy to a) box node.
>  + *  - author: JL
> @@ -9440,7 +9440,7 @@ Index: src/synctex_parser.c
>  +        int bound = 1500000/(node->class->scanner->pre_magnification/1000);
>  +        synctex_node_p parent = NULL;
>  +        /*  get the first enclosing parent
> -+         *  then get the highest enclosing parent with the same mean line 
> ±1 */
> ++         *  then get the highest enclosing parent with the same mean line 
> ±1 */
>  +        node = _synctex_node_or_handle_target(node);
>  +        if (!_synctex_node_is_box(node)) {
>  +            if ((parent = _synctex_tree_parent(node))) {
<#secure method=pgpmime mode=sign>

-- 
jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF  DDCC 0DFA 74AE 1524 E7EE

Reply via email to