On 2014-07-21 23:05-0700 Alan W. Irwin wrote: > On 2014-07-21 21:23+0100 António Rodrigues Tomé wrote: > >> Hi, >> very strange thing. >> latter on I'll test everything >> today i've only changed >> bindings/qt_gui/plqt.cpp >> the way you said and the vertical offsset was smaller so I changed the >> original ofsset signal and the result was a little better. One could easily >> find a new good offset for my qt built, don't know if it is the better >> thing specially if it changes with any minor changes in QT built. >> >> The result with the changes >> // PLFLT empirical_yOffset = -0.63 * chrht * POINTS_PER_INCH / 25.4; >> // PLFLT empirical_yOffset = 0. * chrht * POINTS_PER_INCH / 25.4; >> PLFLT empirical_yOffset = 0.63 * chrht * POINTS_PER_INCH / 25.4; >> >> is attached > > Hi António: > > That was indeed an interesting result, which shows better asterisk > alignment, but worse numerical character alignment. (See how the > numerical labels for the vertical axes are not aligned properly with > the major tick marks). The up side is there are no more tests to do > since those would have only been worthwhile if you had had a perfect > pdf result. > > I think the conclusion must be that vertical alignment issues not only > persist but are much less consistent in 5.3.1 versus 5.2.1 so a simple > Y offset does adjust for the 5.2.1 issues, but cannot deal with the > variety of alignment issues for the asterisk versus numerical > characters your examples shows for 5.3.1. That's disappointing, but > hopefully these Qt5 character alignment issues and also the additional > font issues you showed for the png result will get straightened out, > and meanwhile you should fall back to using the reliable Qt4.
P.S. if you look at <https://bugreports.qt-project.org/issues/?jql=project%20%3D%20QTBUG%20AND%20status%20%3D%20Open%20AND%20text%20~%20%22align%22> (which searches for all open Qt bugs with "align" in their subject), there are 109 hits. Not all of those are relevant to Qt5 text alignment issues, but many of them are so I am not surprised we are encountering some text alignment issues for Qt5. <rant> Also, I am sad to see there is a voting system there for bugs which users feel are important to be fixed. To me that is a strong indication that the Qt5 development team just does not have the resources to pay much attention to bug reports until they receive a lot of votes. Limited development resources are a really sad situation for Qt5. Furthermore, I don't like the priorities that are implied by a voting system. For example, many bug reports are for minor issues (such as text alignment) that should only take ~5 minutes to fix, but which normally would only get a few votes since it is a minor issue. In my view such bug reports for easy-to-fix issues should get extreme priority to (a) greatly increase the perceived overall quality of of Qt5 software due to fixes for the large numbers of minor issues and (b) clear out a lot of the noise in the bug tracker backlog so the remaining more major issues are revealed. In sum, too many projects these days work on bug triage (classification of bugs, assigning bugs, finding duplicates of bugs, and now looking at votes for bugs) as opposed to actually fixing bugs! </rant> Alan __________________________ Alan W. Irwin Astronomical research affiliation with Department of Physics and Astronomy, University of Victoria (astrowww.phys.uvic.ca). Programming affiliations with the FreeEOS equation-of-state implementation for stellar interiors (freeeos.sf.net); the Time Ephemerides project (timeephem.sf.net); PLplot scientific plotting software package (plplot.sf.net); the libLASi project (unifont.org/lasi); the Loads of Linux Links project (loll.sf.net); and the Linux Brochure Project (lbproject.sf.net). __________________________ Linux-powered Science __________________________ ------------------------------------------------------------------------------ Want fast and easy access to all the code in your enterprise? Index and search up to 200,000 lines of code with a free copy of Black Duck Code Sight - the same software that powers the world's largest code search on Ohloh, the Black Duck Open Hub! Try it now. http://p.sf.net/sfu/bds _______________________________________________ Plplot-devel mailing list Plplot-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/plplot-devel
