FS#6588 - Pacman fills terminal with '#' upon resize while downloading

Attached to Project: Arch Linux
Opened by James Rayner (iphitus) - Tuesday, 13 March 2007, 10:10 GMT
Last edited by Dan McGee (toofishes) - Tuesday, 13 March 2007, 16:22 GMT
Task Type Bug Report
Category System
Status Closed
Assigned To Dan McGee (toofishes)
Architecture not specified
Severity Medium
Priority Normal
Reported Version 0.7.2 Gimmick
Due in Version Undecided
Due Date Undecided
Percent Complete 100%
Votes 0
Private No

Details

pacman3-rc2 that phrak posted to the ML today.

Bug: Pacman spams the terminal at a ridiculous rate with '#'. By spam, i mean.... lots, just constant scrolling of '#'. *heaps*. Too much for screen to handle so I had to switch out to another terminal to kill pacman.

To reproduce: I did pacman -S inkscape, then resized the window. If it doesnt the first time, resize afew more and it should go bonkers.

I'm running pacman in screen, with a basic config, just hardstatus + 256color support added. Screen is in an xterm, managed by dwm. If I resize, the terminal while it's downloading packages, it does the above. Doesnt always do it first resize, but it has in the past.

James
This task depends upon

This task blocks these from closing
 FS#6316 - Pacman 3 release bugcatcher 
Closed by  Dan McGee (toofishes)
Wednesday, 14 March 2007, 13:49 GMT
Reason for closing:  Fixed
Additional comments about closing:  Reported as fixed, checked into CVS.
Comment by James Rayner (iphitus) - Tuesday, 13 March 2007, 10:14 GMT
just happened again, but with '-'

and I just changed to a different workspace in dwm, but I suppose there may have been some sort of resize called.
Comment by James Rayner (iphitus) - Tuesday, 13 March 2007, 10:25 GMT
ok... looks like it happens when the terminal is shrunk so small that the progress bar is not visible.

don't know whether the package name makes a difference, but the below i obviously use kernel26suspend2, mainly because it's big and I could copy the progress :)

works
kernel26suspend2 2.7M 184.9K/s 00:01:29 [#######------------------------------------------------] 14%

works ($COLUMNS>=59)
kernel26suspend2 3.9M 1200.2K/s 00:00:12 [###---------------] 20%
kernel26suspend2 2.9M 38.9K/s 00:07:00 [-----] 15%
kernel26suspend2 3.8M 235.1K/s 00:01:05 [-] 20%
kernel26suspend2 4.8M 183.1K/s 00:01:18 [] 25%

wraps ($COLUMNS=58)
"kernel26suspend2 5.7M 111.1K/s 00:02:01 [] 30"
"% "

dies ($COLUMNS<=57)
below this and it just spams - or #

Comment by Dan McGee (toofishes) - Tuesday, 13 March 2007, 14:35 GMT
This shouldn't be a terribly hard fix, I'll take a look at it later today.
Comment by Aaron Griffin (phrakture) - Tuesday, 13 March 2007, 14:59 GMT
Yeah, should be easy.... if maxcols <= infolen + (5?) or so, no progress bar, just spit out info.
Comment by Dan McGee (toofishes) - Tuesday, 13 March 2007, 16:23 GMT
There is a fix checked into pacman CVS, can you try it out for us? I gave it some testing but it is tough with Putty.
Comment by James Rayner (iphitus) - Tuesday, 13 March 2007, 23:56 GMT
sure, will do so when i get home.
Comment by James Rayner (iphitus) - Wednesday, 14 March 2007, 09:10 GMT
seems fixed. good work guys, pacman3 works sweet. Certainly more responsive, and undoubtedly quicker.

James

Loading...