Its been noted before there may be some leak in Nautilus. I bumped
into a possible symptom... 1. Open fresh image (e.g. 50216) 2. Open Transcript. 3. Open *one* System Browser 4. Add instrumentation to method Nautilus >>> NautilusUI >> methodModified: e.g.... Transcript crShow: '>>>' , thisContext printString; tab; show: anAnnouncement printString ; tab ; show: sourceTextModel text. 5. In same System Browser, add method MyPackage >>> MyClass >> test1 Observe steps (4.) and (5.) each logged a single item to Transcript. 6. Open a *second* System Browser and add method MyPackage >>> MyClass >> test2 Observe step (6.) logged two items to Transcript, one for each open System Browser. 7. Close *second* System Browser. 8. In *first* System Browser, modify #test1. Observe step (8.) logged two items to Transcript. I would expect only one. Is there any issue open for this? I couldn't find one. cheers -ben |
> Its been noted before there may be some leak in Nautilus. Yes because probably of announcer. Thanks for reporting them. We started to clean Nautilus and we already simplified a lot some dark corners but there was and still certainly a lot of places for improvements. > I bumped > into a possible symptom... > > 1. Open fresh image (e.g. 50216) > > 2. Open Transcript. > > 3. Open *one* System Browser > > 4. Add instrumentation to method Nautilus >>> NautilusUI >> methodModified: > e.g.... Transcript crShow: '>>>' , thisContext printString; tab; > show: anAnnouncement printString ; tab ; show: sourceTextModel text. > > 5. In same System Browser, > add method MyPackage >>> MyClass >> test1 > > Observe steps (4.) and (5.) each logged a single item to Transcript. > > 6. Open a *second* System Browser and > add method MyPackage >>> MyClass >> test2 > > Observe step (6.) logged two items to Transcript, one for each open > System Browser. > > 7. Close *second* System Browser. > 8. In *first* System Browser, modify #test1. > > Observe step (8.) logged two items to Transcript. I would expect only one. > > > Is there any issue open for this? I couldn't find one. > > cheers -ben > > |
On Tue, Aug 11, 2015 at 3:54 AM, stepharo <[hidden email]> wrote:
> >> Its been noted before there may be some leak in Nautilus. > > > Yes because probably of announcer. Thanks for reporting them. > We started to clean Nautilus and we already simplified a lot some dark > corners but there was and still > certainly a lot of places for improvements. > > >> I bumped >> into a possible symptom... >> >> 1. Open fresh image (e.g. 50216) >> >> 2. Open Transcript. >> >> 3. Open *one* System Browser >> >> 4. Add instrumentation to method Nautilus >>> NautilusUI >> >> methodModified: >> e.g.... Transcript crShow: '>>>' , thisContext printString; tab; >> show: anAnnouncement printString ; tab ; show: sourceTextModel text. >> >> 5. In same System Browser, >> add method MyPackage >>> MyClass >> test1 >> >> Observe steps (4.) and (5.) each logged a single item to Transcript. >> >> 6. Open a *second* System Browser and >> add method MyPackage >>> MyClass >> test2 >> >> Observe step (6.) logged two items to Transcript, one for each open >> System Browser. >> >> 7. Close *second* System Browser. >> 8. In *first* System Browser, modify #test1. >> >> Observe step (8.) logged two items to Transcript. I would expect only >> one. >> >> >> Is there any issue open for this? I couldn't find one. >> >> cheers -ben >> Posts seem to be slow coming through, sorry if I double up this response. Not sure yet if this is all that is required, but its a start... NautilusWindow >> delete | mod | mod := self model. super delete. ">>>" SystemAnnouncer uniqueInstance unsubscribe: mod. "<<<added" mod ifNotNil: [ mod announce: (WindowClosed new window: self) ] cheers -ben |
Free forum by Nabble | Edit this page |