- Pharo-dev - lists.pharo.org

Re: [Pharo-project] FilePlugin changed it's behavior?

by Sean P. DeNigris

Guillermo Polito wrote > > Has this behavior changed lately? > There's something weird and scary going on with FilePlugin... * Issue 6050: ensureDirectory looks strange http://code.google.com/p/pharo/issues/detail?id=6050 * http://forum.world.st/FilePlugin-bug-td4634074.html Anyone else notice anything? - S -- View this message in context: http://forum.world.st/FilePlugin-changed-it-s-behavior-tp4633565p4634255.ht… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

June 11, 2012

Re: [Pharo-project] zip bug [IMPORTANT] probably related to the recent Zip changes

by Sean P. DeNigris

Okay, here is an issue: Issue 6054: Error when reading zip files with Squeak date/time format http://code.google.com/p/pharo/issues/detail?id=6054 with a fix attached and a comment pointing back here: > Attached are two cs files that have a temporary fix enabling us to load > mcz and other zip files, even if they have the old, incorrect date format. > They should be refactored and there is a larger question to discuss [1], > but they work under manual testing. > > The new behavior when reading zip file timestamps is: > * parse as a DOS timestamp > * if an error is encountered, fall back and treat it like a Squeak (i.e. > incorrect) DOS timestamp. This may return an incorrect result, but will > not cause an error > > The changes were tested by hand on Balloon-CamilloBruni.85.mcz, which > Steph was having a problem with because there was 13 as a month index, > which blew up [1] > The larger issue I was thinking about is: should we extend this fallback to all DOS timestamp parsing, but now that I think about it: * I imagine there's not much of that going on besides zipping, lol * I think it'd be better to signal an InvalidDosTimestamp, describing the change, and pointing users to how to convert their data, instead of silently trying to convert What do you think? Cheers, Sean -- View this message in context: http://forum.world.st/zip-bug-IMPORTANT-probably-related-to-the-recent-Zip-… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

June 11, 2012

Re: [Pharo-project] zip bug [IMPORTANT] probably related to the recent Zip changes

by Sean P. DeNigris

Sean P. DeNigris wrote > > let me see write up a fallback defaulting to the old date format for > invalid dates... > Damn!! I'm screwed because the bug we found yesterday in #ensureDirectory (in the VM I think) makes all my tests fail, so I'm refactoring blind!! -- View this message in context: http://forum.world.st/zip-bug-IMPORTANT-probably-related-to-the-recent-Zip-… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

June 11, 2012

Re: [Pharo-project] zip bug [IMPORTANT] probably related to the recent Zip changes

by Sean P. DeNigris

Sean P. DeNigris wrote > > Where did you click on it? I need steps to reproduce... > Okay, I tried to merge it and got the error when the package was being read... I did some experiments... * create a zip at Mac command line -> new Pharo code correctly parses date/times * unzip Balloon-CamilloBruni.85.mcz at Mac command line -> Mac reports date/time as 1/23/2010... I'm assuming this is wrong because pre-fix, Pharo reported 29 September 2011 So, I think the code is correct, but transitioning is a problem... let me see write up a fallback defaulting to the old date format for invalid dates... -- View this message in context: http://forum.world.st/zip-bug-IMPORTANT-probably-related-to-the-recent-Zip-… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

June 11, 2012

Re: [Pharo-project] DBXTalk & seaside / success in Pier one-click / my 1.4 woes follow-up

by Yanni Chiu

On 10/06/12 6:16 PM, Cam S wrote: > Unfortunately, Fuel does not work in this Pier based image, so I still > have the situation where different images allow me to do different > combinations, but not one that allows me to operate my application > completely. (Maybe if i put my head in the sand the problems will go > away... ?) I've never had a problem with Fuel in a Pier image. What problem are you having? (It helps if you are specific in your post; otherwise, people have to ask.) What Pier one-click image did you use (link please)? Was it Magritte2 or Magritte3? How did you load Fuel? Did you use ConfigurationOf? Did you specify #stable or #bleedingEdge?

June 11, 2012

Re: [Pharo-project] zip bug [IMPORTANT] probably related to the recent Zip changes

by Sean P. DeNigris

StÃ©phane Ducasse wrote > > Sean could you have a look? > I can't get latest 2.0 bc Jenkins is down. Can you send me an image? Otherwise I'll load the slice into 20118 StÃ©phane Ducasse wrote > > Because when I tried to click on Balloon-CamilloBruni.85.mcz in PharoInbox > Where did you click on it? I need steps to reproduce... -- View this message in context: http://forum.world.st/zip-bug-IMPORTANT-probably-related-to-the-recent-Zip-… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

June 11, 2012

Re: [Pharo-project] Jenkins 502 errors

by Igor Stasenko

On 11 June 2012 02:59, Sean P. DeNigris <sean(a)clipperadams.com> wrote: > Jenkins seems to be down still... since yesterday? > yupp.. java software. :) we will figure it out at monday > -- > View this message in context: http://forum.world.st/Jenkins-502-errors-tp4634147p4634240.html > Sent from the Pharo Smalltalk mailing list archive at Nabble.com. > -- Best regards, Igor Stasenko.

June 11, 2012

Re: [Pharo-project] Jenkins 502 errors

by Sean P. DeNigris

Jenkins seems to be down still... since yesterday? -- View this message in context: http://forum.world.st/Jenkins-502-errors-tp4634147p4634240.html Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

June 11, 2012

Re: [Pharo-project] zip bug [IMPORTANT] probably related to the recent Zip changes

by Sean P. DeNigris

StÃ©phane Ducasse wrote > > Sean could you have a look? > > Because when I tried to click on Balloon-CamilloBruni.85.mcz in PharoInbox > > I get a month as 13 > I'll take a look. I'm very confident about that code, because I tested it over and over with real data from different sources. Although, it makes sense that, if we've been saving files with wrong values and loading them using the same formula, we wouldn't know they were wrong until we fix the format... If that's the case, maybe wrap the unzip with a check that defaults to the old format if the date is invalid or strange (e.g. year 1962)... Sean -- View this message in context: http://forum.world.st/zip-bug-IMPORTANT-probably-related-to-the-recent-Zip-… Sent from the Pharo Smalltalk mailing list archive at Nabble.com.

June 11, 2012

[Pharo-project] DBXTalk & seaside / success in Pier one-click / my 1.4 woes follow-up

by Cam S

In my quest to get more than one tool working in one image, I built another image. This time I started with a Pier download (today, June 10, 2012) one-click image, and simply attempted load the DBXTalk. Reading straight from the DBXTalk Documentation page (http://dbxtalk.smallworks.com.ar/Documentation) See below. The Pier-based image did not mask the problems I was having in the Pharo1.4 w/Zinc based image, and so I was able to debug the problems related to #asComponent, so that #asComponent works on DBXTalk built classes. Hence the changes to DBXTalk of the former post. However, when the changes are put into my Pharo1.4/zinc/DBXTalk/seaside image, still, nothing is rendered for the "... asComponent addValidatedForm" -- at least I can move forward with my interfaces by using the Pier image. Unfortunately, Fuel does not work in this Pier based image, so I still have the situation where different images allow me to do different combinations, but not one that allows me to operate my application completely. (Maybe if i put my head in the sand the problems will go away... ?) -Cam - The loading script for DBXTalk - Gofer it squeaksource: 'MetacelloRepository'; package: 'ConfigurationOfOpenDBXDriver'; load. (((Smalltalk at: #ConfigurationOfOpenDBXDriver) perform: #project) perform: #version: with: #stable) load. Gofer it squeaksource: 'MetacelloRepository'; package: 'ConfigurationOfGlorpDBX'; load. (((Smalltalk at: #ConfigurationOfGlorpDBX) perform: #project) perform: #version: with: #stable) load. "If you want to use the PostgreSQL smalltalk native driver, you can execute instead:" "(((Smalltalk at: #ConfigurationOfGlorpDBX) perform: #project) perform: #version: with: #stable) load: 'GlorpPostgresV2Native'." Gofer it squeaksource: 'MetacelloRepository'; package: 'ConfigurationOfDBXTools'; load. (((Smalltalk at: #ConfigurationOfDBXTools) perform: #project) perform: #version: with: #stable) load: 'Phoseydon'. "BUT... this err'd so then I manually pulled in most of the DBXTalk packages via monticello. " Gofer it squeaksource: 'MetacelloRepository'; package: 'ConfigurationOfDBXTools'; load. (((Smalltalk at: #ConfigurationOfDBXTools) perform: #project) perform: #version: with: #stable) load: 'Neptuno'

June 10, 2012