[LATE NIGHT PATCH] build gzipped apidoc in case have doxygen but not sphinx

[LATE NIGHT PATCH] build gzipped apidoc in case have doxygen but not sphinx

[Tomi Ollila]

In case we had doxygen but not sphinx notmuch.3 was created but
notmuch.3.gz not -- which means install fails!
This patch (with late night unpolished commit message will fix that)
---

I'll do better one... hmm, next week (unless someone gets there first \o/)

 doc/Makefile.local | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/doc/Makefile.local b/doc/Makefile.local
index 0bdf2e12fc0f..41cba607947d 100644
--- a/doc/Makefile.local
+++ b/doc/Makefile.local
@@ -32,7 +32,9 @@ sphinx-info: sphinx-texinfo
 
 -include $(dir)/docdeps.mk
 
+ifeq ($(HAVE_SPHINX),1)
 MAN_GZIP_FILES := $(addsuffix .gz,${MAN_ROFF_FILES})
+endif
 
 # Use the man page converter that is available. We should never depend
 # on MAN_ROFF_FILES if a converter is not available.
@@ -74,15 +76,14 @@ apidocs:
 install-apidocs:
 endif
 
-# Do not try to build or install man pages if a man page converter is
-# not available.
+build-man: ${MAN_GZIP_FILES}
+install-man: ${MAN_GZIP_FILES}
+
 ifeq ($(HAVE_SPHINX),0)
-build-man:
 install-man:
 	@echo "No sphinx, will not install man pages."
 else
-build-man: ${MAN_GZIP_FILES}
-install-man: ${MAN_GZIP_FILES}
+install-man:
 	mkdir -p "$(DESTDIR)$(mandir)/man1"
 	mkdir -p "$(DESTDIR)$(mandir)/man5"
 	mkdir -p "$(DESTDIR)$(mandir)/man7"
-- 
2.0.0

Re: [LATE NIGHT PATCH] build gzipped apidoc in case have doxygen but not sphinx

[David Bremner]

Tomi Ollila <tomi.ollila@iki.fi> writes:

> In case we had doxygen but not sphinx notmuch.3 was created but
> notmuch.3.gz not -- which means install fails!
> This patch (with late night unpolished commit message will fix that)

By install fails, you mean the man page is simply not installed? I agree
that's a bug, but I had visions of the install failing half way through.

d

Re: [LATE NIGHT PATCH] build gzipped apidoc in case have doxygen but not sphinx

[David Bremner]

David Bremner <david@tethera.net> writes:

> Tomi Ollila <tomi.ollila@iki.fi> writes:
>
>> In case we had doxygen but not sphinx notmuch.3 was created but
>> notmuch.3.gz not -- which means install fails!
>> This patch (with late night unpolished commit message will fix that)
>
> By install fails, you mean the man page is simply not installed? I agree
> that's a bug, but I had visions of the install failing half way through.

A related issue I noticed is that some tests fail, rather than being
skipped, if the man pages are not built.

d

[PATCH] doc: gzipped notmuch.3 before trying to install notmuch.3.gz

[David Bremner]

If HAVE_SPHINX=0 but HAVE_DOXYGEN=1, then the previous version was
trying to install notmuch.3.gz but only got as far as creating
notmuch.3
---

doc/Makefile.local is messy enough that I found Tomi's fix hard to
follow. Probably the real answer is to fix some of this mess, but in the mean time, what about this as an alternative fix?

 doc/Makefile.local | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/doc/Makefile.local b/doc/Makefile.local
index 0bdf2e1..e209749 100644
--- a/doc/Makefile.local
+++ b/doc/Makefile.local
@@ -60,7 +60,7 @@ install-man: install-apidocs
 ifeq ($(HAVE_DOXYGEN),1)
 MAN_GZIP_FILES += ${APIMAN}.gz
 apidocs: $(APIMAN)
-install-apidocs: apidocs
+install-apidocs: ${APIMAN}.gz
 	mkdir -p "$(DESTDIR)$(mandir)/man3"
 	install -m0644  $(DOCBUILDDIR)/man/man3/*.3.gz  $(DESTDIR)/$(mandir)/man3
 
-- 
2.1.4

Re: [PATCH] doc: gzipped notmuch.3 before trying to install notmuch.3.gz

[Tomi Ollila]

On Sun, Jan 25 2015, David Bremner <david@tethera.net> wrote:

> If HAVE_SPHINX=0 but HAVE_DOXYGEN=1, then the previous version was
> trying to install notmuch.3.gz but only got as far as creating
> notmuch.3
> ---
>
> doc/Makefile.local is messy enough that I found Tomi's fix hard to
> follow. Probably the real answer is to fix some of this mess, but in the
> mean time, what about this as an alternative fix?

Sure looks simpler - I also couldn't figure out (just by quickly looking my
patch) how that exactly made things work...

This alternative fix is IMO good for the time being, and fixes a real
problem. I wonder whether gzipped user docs are created at build time but
gzipped apidocs just before install -- I might spare more time checking
that when on my regular linux desktop (these alternatives are just too PITA
to use ;) 

Tomi


>
>  doc/Makefile.local | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/doc/Makefile.local b/doc/Makefile.local
> index 0bdf2e1..e209749 100644
> --- a/doc/Makefile.local
> +++ b/doc/Makefile.local
> @@ -60,7 +60,7 @@ install-man: install-apidocs
>  ifeq ($(HAVE_DOXYGEN),1)
>  MAN_GZIP_FILES += ${APIMAN}.gz
>  apidocs: $(APIMAN)
> -install-apidocs: apidocs
> +install-apidocs: ${APIMAN}.gz
>  	mkdir -p "$(DESTDIR)$(mandir)/man3"
>  	install -m0644  $(DOCBUILDDIR)/man/man3/*.3.gz  $(DESTDIR)/$(mandir)/man3
>  
> -- 
> 2.1.4

Re: [PATCH] doc: gzipped notmuch.3 before trying to install notmuch.3.gz

[David Bremner]

David Bremner <david@tethera.net> writes:

> If HAVE_SPHINX=0 but HAVE_DOXYGEN=1, then the previous version was
> trying to install notmuch.3.gz but only got as far as creating
> notmuch.3
> ---
pushed.

d

Re: [LATE NIGHT PATCH] build gzipped apidoc in case have doxygen but not sphinx

[David Bremner]

David Bremner <david@tethera.net> writes:

> David Bremner <david@tethera.net> writes:
>
>> Tomi Ollila <tomi.ollila@iki.fi> writes:
>>
>>> In case we had doxygen but not sphinx notmuch.3 was created but
>>> notmuch.3.gz not -- which means install fails!
>>> This patch (with late night unpolished commit message will fix that)
>>
>> By install fails, you mean the man page is simply not installed? I agree
>> that's a bug, but I had visions of the install failing half way through.
>
> A related issue I noticed is that some tests fail, rather than being
> skipped, if the man pages are not built.

Both my not understanding Tomi's report and my problem with failing
tests where related to not running "make clean" after removing
sphinx. So I think we can ignore this. Perhaps we should suggest people
run "make clean" whenever they re-run configure, or something along
those lines.

d