Mailman 3 September 2012 - Tkinter-discuss

Control position of Tkinter common dialogs?
by python＠bdurham.com Dec. 9, 2016

Dec. 9, 2016

Python 2.7 under Windows: How can we control the position of Tkinter's common dialogs? Here's what we've discovered: * Certain common dialogs always open up relative to their parent window * Certain common dialogs always open up centered on the user's desktop * All common dialogs appear to ignore the optional parent= parameter Questions: * How can we force a dialog to open up relative to its parent window? * How can we force a dialog to open up centered on the user's desktop? Background: import tkColorChooser as colorchooser import tkFileDialog as filedialog import tkMessageBox as messagebox # always open up relative to parent windows fileOpen = filedialog.askopenfilename() fileOpens = filedialog.askopenfilenames() fileSaveAs = filedialog.asksaveasfilename() color = colorchooser.askcolor() # always open up centered on desktop folderOpen = filedialog.askdirectory() messagebox.askquestion() Malcolm

6 9

Segfault in Image.PhotoImage on OS X
by Marc Culler Nov. 14, 2012

Nov. 14, 2012

This was reported here in February, but no debugging information was provided by the reporter: http://stackoverflow.com/questions/9142786/python-imagetk-photoimage-segfau… <view-source:http://stackoverflow.com/questions/9142786/python-imagetk-photo…> I can explain the nature of the problem, but I don't know how to fix it correctly. I am using the standard x86_64 Python 2.7.2 package from python.org and I have installed Active State Tcl/Tk 8.11. When I tried using easy_install to install PIL it would not load, claiming that _imaging.so did not exist. In fact, it existed but was only a thin i386 executable. So I compiled Imaging-1.1.7 with export CFLAGS='-arch i386 -arch x86_64' and export TCL_ROOT=/Library/Frameworks/Tcl.framework/Versions/8.5/Tcl I was then able to load the module. However, the following code produced a segfault: import Image import ImageTk import Tkinter W = Tkinter.Tk() Im = Image.open('test_image.jpg') x = ImageTk.PhotoImage(Im) The crash occurs in Tk_GetImageMasterData. More importantly, though, the crash was preceded by the following messages: objc[36554]: Class TKApplication is implemented in both /Library/Frameworks/Tk.framework/Versions/8.5/Tk and /System/Library/Frameworks/Tk.framework/Versions/8.5/Tk. One of the two will be used. Which one is undefined. objc[36554]: Class TKMenu is implemented in both /Library/Frameworks/Tk.framework/Versions/8.5/Tk and /System/Library/Frameworks/Tk.framework/Versions/8.5/Tk. One of the two will be used. Which one is undefined. objc[36554]: Class TKContentView is implemented in both /Library/Frameworks/Tk.framework/Versions/8.5/Tk and /System/Library/Frameworks/Tk.framework/Versions/8.5/Tk. One of the two will be used. Which one is undefined. objc[36554]: Class TKWindow is implemented in both /Library/Frameworks/Tk.framework/Versions/8.5/Tk and /System/Library/Frameworks/Tk.framework/Versions/8.5/Tk. One of the two will be used. Which one is undefined. Guessing that PIL was finding the System/Library/Frameworks/Tk.Framework instead of /Library/Frameworks/Tk.Framework, I tried hiding the system Tcl and Tk frameworks by renaming them. I was then able to create the PhotoImage and display it. I don't know where those messages come from, and I have never encountered them before. But I am guessing they are from OS X, since they are generated by an objective C program. Evidently _tkinter.so is able to specify which Tk.Framework to use. I don't know its mechanism for doing that. But, somehow, PIL is not doing it, and it ends up using the wrong one. A stack trace of the crash is below. - Marc 0x000000011517bccd in Tk_GetImageMasterData () (gdb) bt #0 0x000000011517bccd in Tk_GetImageMasterData () #1 0x0000000115183d9f in Tk_FindPhoto () #2 0x00000001014fbadc in PyImagingPhotoPut () #3 0x000000011502a05b in TclInvokeStringCommand () #4 0x000000010102142a in TclEvalObjvInternal () #5 0x000000010102261c in Tcl_EvalObjv () #6 0x000000010100646d in Tkapp_Call () #7 0x00000001000c102d in PyEval_EvalFrameEx () #8 0x00000001000c2d29 in PyEval_EvalCodeEx () #9 0x00000001000c0b6a in PyEval_EvalFrameEx () #10 0x00000001000c2d29 in PyEval_EvalCodeEx () #11 0x000000010003da80 in function_call () #12 0x000000010000c5e2 in PyObject_Call () #13 0x000000010001ebcb in instancemethod_call () #14 0x000000010000c5e2 in PyObject_Call () #15 0x00000001000ba5f7 in PyEval_CallObjectWithKeywords () #16 0x0000000100021e5e in PyInstance_New () #17 0x000000010000c5e2 in PyObject_Call () #18 0x00000001000be5f3 in PyEval_EvalFrameEx () #19 0x00000001000c2d29 in PyEval_EvalCodeEx () #20 0x00000001000c2e46 in PyEval_EvalCode () #21 0x00000001000e769c in PyRun_InteractiveOneFlags () #22 0x00000001000e78fe in PyRun_InteractiveLoopFlags () #23 0x00000001000e80e1 in PyRun_AnyFileExFlags () #24 0x00000001000fe77c in Py_Main () #25 0x0000000100000f14 in dyld_stub_strlen ()

4 5

crash on tkFileDialog:asksaveasfilename()
by Lynn Oliver Sept. 30, 2012

Sept. 30, 2012

I'm puzzled by reports from a single person using my program, who states that he consistently crashes in the common file dialog invoked by asksaveasfilename() if, and only if, he attempts to change folders. He says this is happening on two different machines, an iMac (see below) and a 2011 MacBookPro, both running 10.8.2 (current version of Mountain Lion). The program is using Python 2.7.2, Tcl/Tk 8.5. I have never seen this error or been able to reproduce it on my machine, a 2011 iMac also running 10.8.2. My understanding is that Tkinter is handing control over to a system dialog, so the crash occurs before control returns to Tkinter. Is there anything I could be doing in my program that would affect this? I couldn't find any problem reports similar to this for Tkinter. The crash report he sent is below. Thanks... Lynn Stack Trace: Process: aomDSP [6110] Path: /Applications/aomDSP.app/Contents/MacOS/aomDSP Identifier: aomDSP Version: 1.0.7 (???) Code Type: X86-64 (Native) Parent Process: aomDSP [6108] User ID: 501 Date/Time: 2012-09-29 13:19:05.322 -0400 OS Version: Mac OS X 10.8.2 (12C54) Report Version: 10 Interval Since Last Report: 604288 sec Crashes Since Last Report: 2 Per-App Crashes Since Last Report: 1 Anonymous UUID: 3CA6AC2E-0D37-27A5-22F1-1E1194B4B4EF Crashed Thread: Unknown Exception Type: EXC_BAD_ACCESS (SIGBUS) Exception Codes: KERN_PROTECTION_FAILURE at 0x0000000000000003 Backtrace not available Unknown thread crashed with X86 Thread State (64-bit): rax: 0x00007fff75965860 rbx: 0x0000000000000000 rcx: 0x00007fff76bcde01 rdx: 0x0000000000000007 rdi: 0x00007fff5fbfa790 rsi: 0x00007fff86062c7e rbp: 0x00007fff5fbfa790 rsp: 0x00007fff5fbfa768 r8: 0x00007fff7664c580 r9: 0x0000000000000000 r10: 0x00007fff75965870 r11: 0x00007fff5fbfa634 r12: 0x000000010b6ccd00 r13: 0x000000010b7a72a0 r14: 0x00007fff5fbfa8d8 r15: 0x00007fff5fbfa918 rip: 0x0000000000000003 rfl: 0x0000000000010246 cr2: 0x0000000000000003 Logical CPU: 0 Binary images description not available External Modification Summary: Calls made by other processes targeting this process: task_for_pid: 17 thread_create: 0 thread_set_state: 0 Calls made by this process: task_for_pid: 0 thread_create: 0 thread_set_state: 0 Calls made by all processes on this machine: task_for_pid: 3527 thread_create: 0 thread_set_state: 0 Model: iMac7,1, BootROM IM71.007A.B03, 2 processors, Intel Core 2 Duo, 2.8 GHz, 6 GB, SMC 1.21f4 Graphics: ATI Radeon HD 2600 Pro, ATI,RadeonHD2600, PCIe, 256 MB Memory Module: BANK 0/DIMM0, 2 GB, DDR2 SDRAM, 667 MHz, 0x7F7F7F7F7FF70000, 0x00004B363435365536314535363637460000 Memory Module: BANK 1/DIMM1, 4 GB, DDR2 SDRAM, 667 MHz, 0x7F7FFE0000000000, 0x4542453431554638414244412D36452D4520 AirPort: spairport_wireless_card_type_airport_extreme (0x14E4, 0x88), Broadcom BCM43xx 1.0 (5.10.131.36.16) Bluetooth: Version 4.0.9f33 10885, 2 service, 11 devices, 1 incoming serial ports Network Service: Ethernet, Ethernet, en0 Serial ATA Device: ST31000528AS, 1 TB Parallel ATA Device: MATSHITADVD-R UJ-85J USB Device: Built-in iSight, apple_vendor_id, 0x8502, 0xfd400000 / 3 USB Device: Drobo, 0x19b9, 0x4d10, 0xfd100000 / 2 USB Device: USB 2.0 Hub [MTT], 0x050d (Belkin Corporation), 0x0017, 0xfa400000 / 2 USB Device: Back-UPS BX1300G FW:864.L5 .D USB FW:L5, 0x051d (American Power Conversion), 0x0002, 0xfa440000 / 7 USB Device: My Book 1130, 0x1058 (Western Digital Technologies, Inc.), 0x1130, 0xfa430000 / 6 USB Device: Impulse, 0x1235, 0x001a, 0xfa410000 / 5 USB Device: USB Receiver, 0x046d (Logitech Inc.), 0xc50c, 0xfa460000 / 4 USB Device: My Book 1110, 0x1058 (Western Digital Technologies, Inc.), 0x1110, 0xfa420000 / 3 USB Device: IR Receiver, apple_vendor_id, 0x8242, 0x5d100000 / 2 USB Device: Bluetooth USB Host Controller, apple_vendor_id, 0x8206, 0x1a100000 / 2 FireWire Device: Duet, Apogee Electronics, 400mbit_speed

2 4

Re: [Tkinter-discuss] crash on tkFileDialog:asksaveasfilename()
by Lynn Oliver Sept. 30, 2012

Sept. 30, 2012

He just sent the console output during a crash, so I thought I'd pass that along as well: 2012-09-30 12:41:34.481 PM ReportCrash[935]: Failed to create CSSymbolicatorRef for aomDSP[933] 2012-09-30 12:41:34.589 PM ReportCrash[935]: Failed to create dSYM-less CSSymbolicatorRef for aomDSP[933] 2012-09-30 12:41:34.876 PM ReportCrash[935]: Failed to create CSSymbolicatorRef for aomDSP[930] 2012-09-30 12:41:34.985 PM ReportCrash[935]: Failed to create dSYM-less CSSymbolicatorRef for aomDSP[930] 2012-09-30 12:41:35.099 PM com.apple.launchd.peruser.501[584]: ([0x0-0x4e04e].aomDSP[930]) Job appears to have crashed: Bus error: 10 This looks more like a failure to link to a dynamic library but I'm not sure how to interpret this. Is it a result of the crash, or is it related to an incompatibility that caused the crash? I'm afraid I'm reduced to searching stackoverflow for this type of thing. Lynn

2 1

Re: [Tkinter-discuss] FocusOut not working as expected - tix broken?
by mkieverpy＠tlink.de Sept. 26, 2012

Sept. 26, 2012

Hello Michael, Me: > What I really get is this: I only get FocusOut when the mouse > leaves the application window. I do not get FocusOut event > when I click some other part of the application window > which nonetheless results in the frame losing focus and no longer > receiving key events. > Currently, my guess is that this is broken in tix > (which unluckily seems not to be under development any more). Michael: (...) >and here it worked as expected (python 3.1,, debian squeeze with > IceWM). >It might very well be a WM specific issue though, did you try > different >WMs? Sorry, lost focus completely, so to speak. Hardware failure. Hopefully back next week. Before breakdown I tried IceWM and got same result for my and your program. I'm on debian wheezy, ctwm, self compiled python 3.1.1 (tcl/tk 8.4 not sure) I am not getting any 'focus out' messages from your program. No idea why. I'll try to post a minimal example some time next week. Regards, Matthias. (sent from my mac cookie box)

2 1

FocusOut not working as expected - tix broken?
by mkieverpy＠tlink.de Sept. 25, 2012

Sept. 25, 2012

Dear list, I built a GUI using tkinter and tix (python 3.1/linux) and have a frame which wants to receive key events in a certain "mode". I do this when initializing the frame: class WBuildMenu(Frame, View): ... self._keyID = self.bind('<Key>', self._onKey, add=True) and later when entering the "mode": ... self.focus_set() self._FocusID = self.bind('<FocusOut>' , self._focusLost, add=True) I then receive key events correctly. I expected to get <FocusOut> event when my frame loses focus and no longer receives key events. Am I correct in expecting this behaviour? (I wanted to repeat "self.focus_set()" in this case) What I really get is this: I only get FocusOut when the mouse leaves the application window. I do not get FocusOut event when I click some other part of the application window which nonetheless results in the frame losing focus and no longer receiving key events. Currently, my guess is that this is broken in tix (which unluckily seems not to be under development any more). Any other opinions? Any other method to track focus which I missed? Thanks for any hints, Matthias Kievernagel

2 1

pack geometry manager question
by Lion Kimbro Sept. 19, 2012

Sept. 19, 2012

Dear tkinter-discuss, I'm trying to make a user interface like the following: :::::::::::::::::::::::::::::::::::::::::::: : XXXXXXXXXXXXXX : <- a Label :[ <-- expands ][ expands --> ]: <- 2 Entries :[ <-- expands ][ expands --> ]: <- 2 more Entries : : : (empty space expands : : downward) : :..........................................: Pseudo-code (full code at end of document): toplevel "XXXXXXXXXXXXXX" label anchor=n pack frame expand=1 fill=x anchor=n pack entry expand=1 fill=x anchor=n lpack entry expand=1 fill=x anchor=n lpack frame expand=1 fill=x anchor=n pack entry expand=1 fill=x anchor=n lpack entry expand=1 fill=x anchor=n lpack frame expand=1 fill=both anchor=n pack What I'm finding is this: 1. the label (XXXXXXXXXXXXXX) hugs the top of the window (GOOD) 2. the top two entries hug the label (GOOD) 3. the lower two entries space a ways away from the top two entries (DO NOT WANT) What I want instead, is that the lower two entries hug the top two entries. (More deeply: I want to UNDERSTAND why this is happening like this. I could use a grid and that would solve this problem, but I want to have facility with the packer, too.) I've tried changing around the values a bunch, but I just don't have any luck getting this to work. My assumption (that seems to be flawed) is that when I set expand=1, that if it says "fill=x", then that means that it will ONLY expand horizontally. But it seems to be expanding in both directions. Thoughts? Insights? == Full Example (Python 3.2) == from tkinter import * t = Toplevel() lbl = Label(t, text="XXXXXXXXXXX") lbl.pack(side=TOP, anchor=N) f1 = Frame(t) f1.pack(side=TOP, expand=1, fill=X, anchor=N) e11 = Entry(f1) e11.pack(side=LEFT, expand=1, fill=X, anchor=N) e12 = Entry(f1) e12.pack(side=LEFT, expand=1, fill=X, anchor=N) f2 = Frame(t) f2.pack(side=TOP, expand=1, fill=X, anchor=N) e21 = Entry(f2) e21.pack(side=LEFT, expand=1, fill=X, anchor=N) e22 = Entry(f2) e22.pack(side=LEFT, expand=1, fill=X, anchor=N) # This last one seems to do nothing. # My hope was that it would somehow "pressure" f2 higher, but nope. f3 = Frame(t) f3.pack(side=TOP, expand=1, fill=X, anchor=N)

4 4

DRAG and DROP image
by rhixz06 Sept. 3, 2012

Sept. 3, 2012

hey guys!hmm... I'm kinda new in tkinter...my friends and I were developing an educational game,..I just want to ask if you know an easiest way to drag and drop an image in tkinter.... please I need a reply thanks,.. -- View this message in context: http://python.6.n6.nabble.com/DRAG-and-DROP-image-tp4987204.html Sent from the Python - tkinter-discuss mailing list archive at Nabble.com.

4 4