Nikola build: performs its task randomly?

[Icaro Perseo]

Hi all, these last occasions I noticed a strange behavior in nikola build as if it ran erratically or randomly because sometimes you need to repeat the process several times to indicate that all outstanding tasks are completed successfully.

Below I show why:

.  render_sources:output/stories/gnulinux-es-dificil/index.md
.  render_posts:cache/stories/gnulinux-es-dificil.html.en
.  render_posts:cache/stories/gnulinux-es-dificil.html
.  gzip:render_sources:output/stories/gnulinux-es-dificil/index.md.gz
.  render_pages:output/en/stories/gnulinux-is-difficult/index.html
.  render_pages:output/stories/gnulinux-es-dificil/index.html
.  gzip:render_pages:output/en/stories/gnulinux-is-difficult/index.html.gz
.  gzip:render_pages:output/stories/gnulinux-es-dificil/index.html.gz
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
(nikola)user@user-Satellite:~/Documentos/perseosblog$ ./build
Scanning posts..done!
.  gzip:render_pages:output/stories/gnulinux-es-dificil/index.html.gz
.  gzip:render_pages:output/en/stories/gnulinux-is-difficult/index.html.gz
.  sitemap:output/sitemap.xml
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
(nikola)user@user-Satellite:~/Documentos/perseosblog$ ./build
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!
Scanning posts..done!

./buid is a small bash script that the only thing it does is nikola build multiply by 9 times. Why? Precisely because apparently I've noticed a build nikola or two is not enough to make sure everything has been completed well, but after this I think it's ridiculous, I joked around with nikola build * 9 but it turns out even that is not sufficient :(.

[Roberto Alsina]

On 13/05/13 17:58, Icaro Perseo wrote:

./buid is a small bash script that the only thing it does is nikola build multiply by 9 times. Why? Precisely because apparently I've noticed a build nikola or two is not enough to make sure everything has been completed well, but after this I think it's ridiculous, I joked around with nikola build * 9 but it turns out even that is not sufficient :(.

There was a bug like that in master but it has been fixed with this:

https://github.com/ralsina/nikola/commit/0dc54997b3c75cefe0b23ba37b0f7fa524f0c106

There is also a problem with python 3, where it never believes it's done, but I have had no success figuring it out yet:

https://github.com/ralsina/nikola/issues/495

[Icaro Perseo]

The catch is that I use the master code :(

user@user-Satellite:~/Documentos/nikola$ git pull
Already up-to-date.

and I have applied all the changes in my system :(

[Roberto Alsina]

On 13/05/13 18:30, Icaro Perseo wrote:
> The catch is that I use the master code :(
>
> user@user-Satellite:~/Documentos/nikola$ git pull
> Already up-to-date.
>
> and I have applied all the changes in my system :(
>

Python 2 or python 3?

[Icaro Perseo]

Sorry, My previous message appeared as deleted :(

Python 2.7...

El lunes, 13 de mayo de 2013 15:58:52 UTC-5, Icaro Perseo escribió:

[Roberto Alsina]

On 14/05/13 15:04, Icaro Perseo wrote:

Sorry, My previous message appeared as deleted :(

Python 2.7...

Doesn't happen here at all. Could you add "nikola dumpdb" between runs and send me the log?

--
You received this message because you are subscribed to the Google Groups "nikola-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to nikola-discus...@googlegroups.com.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[Roberto Alsina]

On 14/05/13 15:04, Icaro Perseo wrote:

Sorry, My previous message appeared as deleted :(

Python 2.7...

Doesn't happen here at all. Could you add "nikola dumpdb" between runs and send me the log?

--

[Icaro Perseo]

Here is the log. If you need anything else, please let me know :).

El lunes, 13 de mayo de 2013 15:58:52 UTC-5, Icaro Perseo escribió:

[Icaro Perseo]

If you need anything else, please let me know :).

El lunes, 13 de mayo de 2013 15:58:52 UTC-5, Icaro Perseo escribió:

[Brad Miller]

I am seeing some similar behavior with the latest HEAD.

I have about 180 blog posts, some of them are quite old, dating back to 2009, 10, and 11.  there is a group of them that seem to be constantly rebuilt.  Not every time, but in a recurring way, maybe every time, maybe every second time.  If I type nikola build a bunch of times, 

I'm using Python 3.4.

I've tried dumping my build db, but I get:

DBM type is 'dbm.ndbm'

ERROR: ndbm does not support iteration of elements

I've tried switching to the sqlite3 backend, but then I just get an error that the database type cannot be determined.

I would be happy to dig in and help track this down, if someone could point me toward some things I could try.

Brad

[Brad Miller]

Just a followup to my previous post, this does appear to be a Python3 issue, because I built a new virtual environment with 2.7 and now only the things that need to build get built.

Brad

[Chris “Kwpolska” Warrick]

On Wed, Apr 23, 2014 at 11:52 PM, Brad Miller <mill...@luther.edu> wrote:
> I am seeing some similar behavior with the latest HEAD.
>
> I have about 180 blog posts, some of them are quite old, dating back to
> 2009, 10, and 11. there is a group of them that seem to be constantly
> rebuilt. Not every time, but in a recurring way, maybe every time, maybe
> every second time. If I type nikola build a bunch of times,
>
> I'm using Python 3.4.
>
> I've tried dumping my build db, but I get:
>
> DBM type is 'dbm.ndbm'
> ERROR: ndbm does not support iteration of elements
>
> I've tried switching to the sqlite3 backend, but then I just get an error
> that the database type cannot be determined.
>
> I would be happy to dig in and help track this down, if someone could point
> me toward some things I could try.

What are the filenames (input and output) and slugs of the affected
posts? This might be an encoding problem.

Also, could you try dumping the dependency database with the `dbm`
backend? Or the `json` backend?

--
Chris “Kwpolska” Warrick <http://kwpolska.tk>
PGP: 5EAAEA16
stop html mail | always bottom-post | only UTF-8 makes sense

[Brad Miller]

On Thursday, April 24, 2014 8:25:19 AM UTC-5, Chris Warrick wrote:

On Wed, Apr 23, 2014 at 11:52 PM, Brad Miller <mill...@luther.edu> wrote:
> I am seeing some similar behavior with the latest HEAD.
>
> I have about 180 blog posts, some of them are quite old, dating back to
> 2009, 10, and 11.  there is a group of them that seem to be constantly
> rebuilt.  Not every time, but in a recurring way, maybe every time, maybe
> every second time.  If I type nikola build a bunch of times,
>
> I'm using Python 3.4.
>
> I've tried dumping my build db, but I get:
>
> DBM type is 'dbm.ndbm'
> ERROR: ndbm does not support iteration of elements
>
> I've tried switching to the sqlite3 backend, but then I just get an error
> that the database type cannot be determined.
>
> I would be happy to dig in and help track this down, if someone could point
> me toward some things I could try.

What are the filenames (input and output) and slugs of the affected
posts?  This might be an encoding problem.

The file names are widely varied, from as simple as:

file --mime-encoding  joshua-tree-national-park.md

joshua-tree-national-park.md: utf-8

(nikola27)bmiller@zebra ~/t/r/p/2/03 (master)>

head joshua-tree-national-park.md

<!--

.. title: Joshua Tree National Park

.. date: 2011/03/24

.. slug: joshua-tree-national-park

.. tags: Travel

.. link:

.. description:

-->

to

file --mime-encoding here-s-a-picture-of-my-new-friends-in-the-computer-science-class-at-oshigambo-hi.md

here-s-a-picture-of-my-new-friends-in-the-computer-science-class-at-oshigambo-hi.md: utf-8

(nikola27)bmiller@zebra ~/t/r/p/2/06 (master)>

head here-s-a-picture-of-my-new-friends-in-the-computer-science-class-at-oshigambo-hi.md

<!--

.. title: Computer Science Class

.. date: 2012/06/23

.. slug: here-s-a-picture-of-my-new-friends-in-the-computer-science-class-at-oshigambo-hi

.. tags: Namibia

.. link:

.. description:

-->

 

Also, could you try dumping the dependency database with the `dbm`
backend?  Or the `json` backend?

When I try to dump the dbm database I get the error:  ndbm does not support iteration of elements   when I tried using a sqlite3 backed dumping failed altogether.  I suppose I could just do a big select dump, or send you the sqlite file.

Brad