threads / patch / 23559

patchDocumentation/Makefile: fix interrupted builds of user-manual.xml

Subject: [PATCH] Documentation/Makefile: fix interrupted builds of user-manual.xml

## tl;dr

3 messages between Apr 22, 2010 and Apr 22, 2010. Diffs are folded; open one to read it.

replies: 2people: 2as markdown or json

Jonathan Nieder· Apr 22, 2010, 01:18 UTC · lore

Unlike gcc, asciidoc does not atomically write its output file or delete it when interrupted. If it is interrupted in the middle of writing an XML file, the result will be truncated input for xsltproc.

	XSLTPROC user-manual.html
	user-manual.xml:998: parser error : Premature end of data in t

Take care of this case by writing to a temporary and renaming it when finished.

Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
---
Based on a true story.
 Documentation/Makefile |    8 ++++++--
 1 files changed, 6 insertions(+), 2 deletions(-)
Show changes to Documentation/Makefile +6 −2
diff --git a/Documentation/Makefile b/Documentation/Makefile
index 8a8a395..04f69cf 100644
--- a/Documentation/Makefile
+++ b/Documentation/Makefile
@@ -264,7 +264,9 @@ manpage-base-url.xsl: manpage-base-url.xsl.in
 	mv $@+ $@
 
 user-manual.xml: user-manual.txt user-manual.conf
-	$(QUIET_ASCIIDOC)$(ASCIIDOC) $(ASCIIDOC_EXTRA) -b docbook -d book $<
+	$(QUIET_ASCIIDOC)$(RM) $@+ $@ && \
+	$(ASCIIDOC) $(ASCIIDOC_EXTRA) -b docbook -d book -o $@+ $< && \
+	mv $@+ $@
 
 technical/api-index.txt: technical/api-index-skel.txt \
 	technical/api-index.sh $(patsubst %,%.txt,$(API_DOCS))
@@ -278,7 +280,9 @@ XSLT = docbook.xsl
 XSLTOPTS = --xinclude --stringparam html.stylesheet docbook-xsl.css
 
 user-manual.html: user-manual.xml
-	$(QUIET_XSLTPROC)xsltproc $(XSLTOPTS) -o $@ $(XSLT) $<
+	$(QUIET_XSLTPROC)$(RM) $@+ $@ && \
+	xsltproc $(XSLTOPTS) -o $@+ $(XSLT) $< && \
+	mv $@+ $@
 
 git.info: user-manual.texi
 	$(QUIET_MAKEINFO)$(MAKEINFO) --no-split -o $@ user-manual.texi

-- 
1.7.1.rc1
Chris Johnsen· Apr 22, 2010, 06:13 UTC · re: Jonathan Nieder · lore

Re: [PATCH] Documentation/Makefile: fix interrupted builds of user-manual.xml

(Resend, including the all original recipients. Sorry for the dupe, Jonathan.)

On 2010 Apr 21, at 20:18, Jonathan Nieder wrote:
Show 14 quoted lines
> Unlike gcc, asciidoc does not atomically write its output file or
> delete it when interrupted.  If it is interrupted in the middle of
> writing an XML file, the result will be truncated input for xsltproc.
>
> 	XSLTPROC user-manual.html
> 	user-manual.xml:998: parser error : Premature end of data in t
>
> Take care of this case by writing to a temporary and renaming it when
> finished.
>
> Signed-off-by: Jonathan Nieder <jrnieder@gmail.com>
> ---
> Based on a true story.
>

For what it is worth, the changes seem reasonable to me. There may be a few other places that could use the same treatment, though maybe not all the other documentation tools suffer to the same degree as asciidoc.

Show 12 quoted lines
> @@ -278,7 +280,9 @@ XSLT = docbook.xsl
>  XSLTOPTS = --xinclude --stringparam html.stylesheet docbook-xsl.css
>
>  user-manual.html: user-manual.xml
> -	$(QUIET_XSLTPROC)xsltproc $(XSLTOPTS) -o $@ $(XSLT) $<
> +	$(QUIET_XSLTPROC)$(RM) $@+ $@ && \
> +	xsltproc $(XSLTOPTS) -o $@+ $(XSLT) $< && \
> +	mv $@+ $@
>
>  git.info: user-manual.texi
>  	$(QUIET_MAKEINFO)$(MAKEINFO) --no-split -o $@ user-manual.texi
>

I suppose it is obvious enough what is going on in this in this second hunk, but there is no mention of this change in the commit message. Is this hunk due to a similar problem with xsltproc? Or maybe this is just "defensive coding"?

-- 
Chris
Jonathan Nieder· Apr 22, 2010, 10:24 UTC · re: Chris Johnsen · lore

Re: [PATCH] Documentation/Makefile: fix interrupted builds of user-manual.xml

Chris Johnsen wrote:
> On 2010 Apr 21, at 20:18, Jonathan Nieder wrote:
Show 6 quoted lines
>> user-manual.html: user-manual.xml
>> -	$(QUIET_XSLTPROC)xsltproc $(XSLTOPTS) -o $@ $(XSLT) $<
>> +	$(QUIET_XSLTPROC)$(RM) $@+ $@ && \
>> +	xsltproc $(XSLTOPTS) -o $@+ $(XSLT) $< && \
>> +	mv $@+ $@
>>
[...]
> I suppose it is obvious enough what is going on in this in this
> second hunk, but there is no mention of this change in the commit
> message. Is this hunk due to a similar problem with xsltproc? Or
> maybe this is just "defensive coding"?

xsltproc indeed has the same problem, as I learned while testing the first hunk. Thanks for asking.

Sorry for the incomplete explanation before. Jonathan

← back to recent threads