Are you using docbook xsl 1.72? There are known problems building the
manpages with that version. 1.71 works, and 1.73 should work when it get
released.
No, I am using docbook-style-xsl-1.68.1-1, and compiling git from RPM.
Besids, I have different set of problems; only with literal blocks.
So while DOCBOOK_XSL_172 is not set... let me check...
Hmmm...
Before
$ man git-diff-tree
Example:
.ft C
:100644 100644 5be4a4...... 000000...... M file.c
.ft
$ make DOCBOOK_XSL_172=YesPlease doc
$ man Documentation/git-diff-tree.1
Example:
:100644 100644 5be4a4...... 000000...... M file.cWhen -z option is
not used, TAB, LF, and backslash characters in pathnames are repre-
sented as \t, \n, and \\, respectively.
So setting DOCBOOK_XSL_172 while it fixes the bug (even if docbook-xsl
is version 1.68.1, not 1.72), but introduces another.
Unfortunately I don't know enough about AsciiDoc, DocBook and XML
toolchain to even *try* to fix this issue.
P.S. Alternate soution would be for SRPM to contain pre-compiled
manpages, and not require asciidoc and its toolchain to provide
manpages.
--
Jakub Narebski
Poland
Are you using docbook xsl 1.72? There are known problems building the
manpages with that version. 1.71 works, and 1.73 should work when it get
released.
I was able to solve this problem with this patch, which adds a XSL file
used specifically for DOCBOOK_XSL_172=YesPlease and where dots and
backslashes are escaped properly so they won't be substituted to the
wrong thing further down the "DocBook XSL pipeline". Doing the escaping
in the existing callout.xsl breaks v1.70.1. Hopefully v1.73 will end
this part of the manpage nightmare.
@@ -0,0 +1,17 @@+<!-- callout.xsl: converts asciidoc callouts to man page format -->+<xsl:stylesheetxmlns:xsl="http://www.w3.org/1999/XSL/Transform"version="1.0">+<xsl:templatematch="co">+<xsl:value-ofselect="concat('▓fB(',substring-after(@id,'-'),')▓fR')"/>+</xsl:template>+<xsl:templatematch="calloutlist">+<xsl:text>⌂sp </xsl:text>+<xsl:apply-templates/>+<xsl:text> </xsl:text>+</xsl:template>+<xsl:templatematch="callout">+<xsl:value-ofselect="concat('▓fB',substring-after(@arearefs,'-'),'. ▓fR')"/>+<xsl:apply-templates/>+<xsl:text>⌂br </xsl:text>+</xsl:template>++</xsl:stylesheet>
Besids, I have different set of problems; only with literal blocks.
So while DOCBOOK_XSL_172 is not set... let me check...
Hmmm...
Before
$ man git-diff-tree
Example:
.ft C
:100644 100644 5be4a4...... 000000...... M file.c
.ft
$ make DOCBOOK_XSL_172=YesPlease doc
$ man Documentation/git-diff-tree.1
Example:
:100644 100644 5be4a4...... 000000...... M file.cWhen -z option is
not used, TAB, LF, and backslash characters in pathnames are repre-
sented as \t, \n, and \\, respectively.
So setting DOCBOOK_XSL_172 while it fixes the bug (even if docbook-xsl
is version 1.68.1, not 1.72), but introduces another.
Unfortunately I don't know enough about AsciiDoc, DocBook and XML
toolchain to even *try* to fix this issue.
I have no idea what this could be. The literallayout block is
overwritten in Documentation/asciidoc.conf unless when DOCBOOK_XSL_172
is set. Newer AsciiDoc installations seem to use a different set of
DocBook tags than what is overwritten by git's asciidoc.conf.
--
Jonas Fonseca
Are you using docbook xsl 1.72? There are known problems building the
manpages with that version. 1.71 works, and 1.73 should work when it get
released.
I was able to solve this problem with this patch, which adds a XSL file
used specifically for DOCBOOK_XSL_172=YesPlease and where dots and
backslashes are escaped properly so they won't be substituted to the
wrong thing further down the "DocBook XSL pipeline". Doing the escaping
in the existing callout.xsl breaks v1.70.1. Hopefully v1.73 will end
this part of the manpage nightmare.
I have applied this patch, and it makes manpages worse, not better.
I use DocBook XSL version 1.68.1
[...]
quoted
Besids, I have different set of problems; only with literal blocks.
So while DOCBOOK_XSL_172 is not set... let me check...
Hmmm...
Before
$ man git-diff-tree
Example:
.ft C
:100644 100644 5be4a4...... 000000...... M file.c
.ft
$ make DOCBOOK_XSL_172=YesPlease doc
$ man Documentation/git-diff-tree.1
Example:
:100644 100644 5be4a4...... 000000...... M file.cWhen -z option is
not used, TAB, LF, and backslash characters in pathnames are repre-
sented as \t, \n, and \\, respectively.
So setting DOCBOOK_XSL_172 while it fixes the bug (even if docbook-xsl
is version 1.68.1, not 1.72), but introduces another.
Unfortunately I don't know enough about AsciiDoc, DocBook and XML
toolchain to even *try* to fix this issue.
I have no idea what this could be. The literallayout block is
overwritten in Documentation/asciidoc.conf unless when DOCBOOK_XSL_172
is set. Newer AsciiDoc installations seem to use a different set of
DocBook tags than what is overwritten by git's asciidoc.conf.
Now I have in git-diff-tree(1), result of (after applying above patch)
$ man Documentation/git-diff-tree.1
git-diff-files [<pattern>...]
compares the index and the files on the filesystem.
An output line is formatted this way:
in-place edit :100644 100644 bcd1234... 0123456... M file0
copy-edit :100644 100644 abcd123... 1234567... C68 file1 file2
rename-edit :100644 100644 abcd123... 1234567... R86 file1 file3
create :000000 100644 0000000... 1234567... A file4 delete
:100644 000000 1234567... 0000000... D file5 unmerged :000000
000000 0000000... 0000000... U file6That is, from the left to the
right:
[...]
Example:
:100644 100644 5be4a4...... 000000...... M file.cWhen -z option is
not used, TAB, LF, and backslash characters in pathnames are repre-
sented as \t, \n, and \\, respectively.
Compare this to output of "man git-diff-tree", same situation, compiled
from SRPM, without any aditional options:
git-diff-files [<pattern>...]
compares the index and the files on the filesystem.
An output line is formatted this way:
.ft C
in-place edit :100644 100644 bcd1234... 0123456... M file0
copy-edit :100644 100644 abcd123... 1234567... C68 file1 file2
rename-edit :100644 100644 abcd123... 1234567... R86 file1 file3
create :000000 100644 0000000... 1234567... A file4
delete :100644 000000 1234567... 0000000... D file5
unmerged :000000 000000 0000000... 0000000... U file6
.ft
That is, from the left to the right:
[...]
Example:
.ft C
:100644 100644 5be4a4...... 000000...... M file.c
.ft
When -z option is not used, TAB, LF, and backslash characters in pathnames
are represented as \t, \n, and \\, respectively.
HTH.
P.S. writing subset of AsciiDoc in Perl, which would write manpages, HTML
and perhaps info/texinfo files directly, without fragile xmlto toolchain,
looks better and better...
$ asciidoc --version
asciidoc 7.1.2
$ rpm -qa docbook*
docbook-utils-0.6.14-4
docbook-style-xsl-1.68.1-1
docbook-style-dsssl-1.79-1
docbook-utils-pdf-0.6.14-4
docbook-dtds-1.0-26
$ xmlto --version
xmlto version 0.0.18
--
Jakub Narebski
Poland
Are you using docbook xsl 1.72? There are known problems building the
manpages with that version. 1.71 works, and 1.73 should work when it get
released.
I was able to solve this problem with this patch, which adds a XSL file
used specifically for DOCBOOK_XSL_172=YesPlease and where dots and
backslashes are escaped properly so they won't be substituted to the
wrong thing further down the "DocBook XSL pipeline". Doing the escaping
in the existing callout.xsl breaks v1.70.1. Hopefully v1.73 will end
this part of the manpage nightmare.
I have applied this patch, and it makes manpages worse, not better.
I use DocBook XSL version 1.68.1
OK, I might have been a bit unclear, but the patch was not intended for
you but for users of version 1.72+. ;)
P.S. writing subset of AsciiDoc in Perl, which would write manpages, HTML
and perhaps info/texinfo files directly, without fragile xmlto toolchain,
looks better and better...
But not as easy as just pulling the documentation branches.
--
Jonas Fonseca
From: Jakub Narebski <hidden> Date: 2016-06-15 22:44:10
Jakub Narebski wrote:
I have applied this patch, and it makes manpages worse, not better.
I use DocBook XSL version 1.68.1
Sorry, I must have had run "make doc" without "make clean" first.
Of course patch doesn't change anything for pre 1.72, and marked
as such.
One again, I'm sory for the confusion.
The old .ft C... .ft error remains.
--
Jakub Narebski
Poland