It has been almost a year since 6.x-1.18 was released; we have seen a lot of bugs fixed and improvements since then. It would be nice to have a new stable version out soon - to get several bugs fixed I have been running the dev version for a while now, and at least from my experience it is much better than 6.x-1.18.

Comments

mikeytown2’s picture

mikeytown2 CreditAttribution: mikeytown2 commented

Not getting a new version out for some time has been bugging me.

Most important thing that needs to happen for a release: Write out a very clear explanation for the newer htaccess rules. I split them up so there is now an htaccess file in the cache dir. The old way will work, but going this route should be slightly faster. Always looking for more speed.

Still chasing bugs; currently finding them in views
#988680: Memory leak with $view->pre_execute() & $view->attach_displays()
#1001542: $view->set_display($display_id); returns TRUE for displays that do not exist.

Still have a couple show stoppers from my point of view:
http://drupal.org/project/issues/search/boost?issue_tags=1.19+Release+Bl...
Biggest one is getting it working with Javascript aggregator's minify again.

I also need to review what's in our local SVN and whats in drupal's CVS; generalize and push code out if it's easily done.

Anonymous’s picture

Anonymous (not verified) CreditAttribution: Anonymous commented

Thanks for your very fast response. And for all your hard work on this. I really appreciate it. I will keep running the latest dev versions and report back on anything I find - so far it all seems to run smoothly. More speed and less memory usage is of course always very welcome!

Manuel Garcia’s picture

Manuel Garcia CreditAttribution: Manuel Garcia commented

Any chances of this happening any time soon? We are considering switching to the dev branch, but it scares us quite a bit to do so...

NITEMAN’s picture

NITEMAN CreditAttribution: NITEMAN commented
Priority: Normal » Major

4 months have passed, could you please review/re-evalue the possibility?

Best Regards

j0nathan’s picture

j0nathan CreditAttribution: j0nathan commented

+1 for a new stable release.
Subscribing.

anarcat’s picture

anarcat CreditAttribution: anarcat commented

+1 here. This is a must, release early, release often!

Anonymous’s picture

Anonymous (not verified) CreditAttribution: Anonymous commented

To be fair, Boost has followed the "release early, release often" policy. I have not seen many modules where bug reports so quickly lead to a fix and a new release. However, this all happens in the dev-version, that is now so much better than the last stable that everyone not updating is missing out hugely. Maybe, if dev is seen as still needing too much work, perhaps we can get a new alpha - so those of us using Boost on production sites can move to that and stay with it until another particularly stable version becomes alpha2 and then eventually beta and stable. Like many, I am now stuck in dev and pretty much feel I need to do every upgrade as it is sometimes difficult to tell whether a dev upgrade fixes a bug or introduces new features, so getting at least an alpha would help a lot!

marios88’s picture

marios88 CreditAttribution: marios88 commented

+1

harrrrrrr’s picture

harrrrrrr CreditAttribution: harrrrrrr commented

any news on this?

Manuel Garcia’s picture

Manuel Garcia CreditAttribution: Manuel Garcia commented
Status: Active » Needs review

Looking at the release blocking issues, it seems to me that there's nothing that isn't already a problem in the current stable release, I'm I wrong? If not, can we just make another release when those are fixed, but let everyone benefit from the numerous bugs fixed already in dev?

mikeytown2’s picture

mikeytown2 CreditAttribution: mikeytown2 commented

The crawler is not in a good state, it needs to be fixed before a new release can be made.

rnyberg’s picture

rnyberg CreditAttribution: rnyberg commented

Yup, the Crawler is a bugnest at best, and I'm not saying it to bash your hard efforts, it was totally messed up for our i18n site. At first I thought it was due heavy modifications to Drupal, but then we realized it was just lacking support for large scale, multidomain translated site. (Good example was if the crawler crawled the frontpage in english, fr.site.com people saw an english frontpage).

On of the reason for the bugging is that each language doesn't have it's own nodes, but all nodes are dynamic depending on language.

For now our production site's crawler is manually built using cURL and recursive self-calling to create 4 threads (10k nodes per thread), and a 1 sec timeout. Kinda mimicing anonymous visitors landing on the pages, the URL arrays are built from node languages and path aliases. It does a pretty ok job of caching our 40k nodes in two to four hours, depending on what else is happending on the server.

I'm yapping cause I wanted to subscribe aswell :)

bgm’s picture

bgm CreditAttribution: bgm commented

Could it be possible to split the Crawler into a separate sub-module? It is a useful feature, but is specific to a some use-cases (imho) and people tend to switch it on without being aware of the impact it may have on the server (we have had a few overload issues in our shared hosting because of buggy sites being crawled endlessly).

Having it in a sub-module could also allow it to have a hook system that other modules could use to do checks on a website (broken links, SEO tests, unit tests, who knows..).

Although, to be honest, I am mostly looking forward to a new release, it is blocking a few issues for provision_boost (drush integration, notably).

Peter Bowey’s picture

Peter Bowey CreditAttribution: Peter Bowey commented

+1 on this!

Well for me, various issues with the following Drupal projects; 'boost', 'advagg', 'cdn' and 'css_emimage' have caused me to look elsewhere for D6 answers... (sigh)!

I have returned to a 'commercial $$ project' that I used 2 years past with Wordpress + Joomla ->
http://code.google.com/p/web-optimizator/

The above project does have a free $0.00 element - but it reduces the usefulness. I bought it commercially 2 years past; and yes it works with Drupal (D6 & D7), Joomla, Wordpress and several other CMS systems.

I am 'still' amazed (fazed) that even 'humble' Wordpress has better + more CDN and Aggregation asset tools than Drupal!!

Where and what are the 'clinch' professional CMS programmers doing?? Duh!