Hello folks, Just one thing about the topic: we do have all sorts of unit testing... just check http://wiki.documentfoundation.org
Best, Charles. Le Sun, 25 Jan 2015 13:12:27 -0800, Spencer Graves <[email protected]> a écrit : > Hi, Frank: > > > Thanks for your clarification. I agree on nearly all > counts. > > > What do LO developers do regarding unit testing — automated > tests to confirm that certain features continue to work? I > contribute to another free open-source software (FOSS) project > (r-project.org <http://r-project.org/>), where “trustworthy software” > is ensured (when it is) by a process of developing test suites and > documentation in parallel with the code. Ideally, we’d like to have > 100% coverage of all documented features. Then any fix of one bug > that broke something else would be flagged in the next test cycle. > However, this is (as you say) a volunteer project, and the coverage > is never 100%. I don’t take the time to write tests for every > feature, but when new bugs are reported, I add tests for those before > I actually fix them. And some organizations pay people full time to > support FOSS projects. (I’ve heard that Google supports Linux to > ensure that they get features and bug fixes they want. LO would be > better if more large organizations make similar investments in LO.) > > > Best Wishes > Spencer > > > > On Jan 25, 2015, at 12:30 PM, CVAlkan <[email protected]> wrote: > > > > Hi Spenser: > > > > What you're saying is certainly true, but in this case it really > > didn't sound like it was "pro-Microsoft" but rather "anti-LO" - a > > subtle difference perhaps, but I believe a significant one. > > > > The real danger, I suspect, is that, knowing we're gaining on the > > proprietary world, we tend to automatically discount some of the > > comments that are put out assuming they are "sour grapes" or > > marketing misinformation. In particular, however, the one > > responder's comment: "contains lot of regression errors in the most > > simple things" (I'm assuming the writer means "basic" as opposed to > > "simple") is really not that difficult to justify over the course > > of the 4.x releases, and not recognizing that will not result in > > the continued progress I think we all want. Even realizing that the > > 4.x series represents a major step forward in lots of ways, there > > are quite a surprising number of things that seem to have been > > "broken" along the path to cleaning and enhancing the code. > > > > I realize that "quite a surprising number of things" is a little > > vague, so I'll mention that one piddly insignificant user alone who > > isn't in any way involved in development or testing (that would be > > me) has filed several bugs that can be looked up: e.g. 74056, > > 86578, 88208, and all of these were for things that I recall > > successfully using in the past. And it isn't hard to locate other > > new bugs related to indexing, printing, table formatting and so > > forth that others have filed. Whether these bugs were introduced > > due to over-enthusiastic coding, "cowboy coding" (as we used to > > call it, a flawed integration process, lack of testing, lack of > > code reviews, and so forth is, of course, not for me to say. And, > > I'm using the phrase "lack of testing" to include an amateurish > > reliance on simple "does it work?" testing as opposed to doing > > "real" testing, which can be summed up as "does everything else > > still work?" (that's why good testers are a phenomenally underrated > > bunch!) > > > > I suspect that's one of the difficulties inherent in this sort of > > development environment: the work is more or less voluntary; > > developing code is fun and results in creative satisfaction > > (positive feedback); rigorous testing for most is not (the best > > result is neutral and often seems or is viewed as unproductive). > > > > One other thing to realize is that the comments I pointed to seem > > to be specific to Writer, which is where the current threat to the > > proprietary world lies (so far as I can tell, Calc has already > > reached sufficient parity with Excel for most typical users) and > > where the naysayers you refer to are taking their current stand. > > > > But, sadly, other modules (particularly Impress) suffer from too > > much attention to cool new features (some of which are admittedly > > nice), and insufficient attention to serious flaws in fundamental > > functionality, like remaining on the same slide when you switch > > views, handling of tables, and so forth. > > > > I'll reiterate the purpose of my post, though: I just thought folks > > should be aware of these sorts of postings, since only LibreOffice > > users (and not potential ones) are likely to be on this particular > > forum (hence - we're preaching to the choir as they say). If a > > major multiple operating system forum has this sort of posting that > > goes unchallenged, that's another matter entirely; my thought was > > that even a public offer of help - specific questions and so forth > > - would be a positive counter-balance. > > > > Frank > > > > > > > > -- > > View this message in context: > > http://nabble.documentfoundation.org/Interesting-comment-about-Writer-Bugs-in-LinuxQuestions-org-tp4137415p4137497.html > > Sent from the Users mailing list archive at Nabble.com. > > > > -- > > To unsubscribe e-mail to: [email protected] > > Problems? > > http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/ > > Posting guidelines + more: > > http://wiki.documentfoundation.org/Netiquette List archive: > > http://listarchives.libreoffice.org/global/users/ All messages sent > > to this list will be publicly archived and cannot be deleted > > > > -- Charles-H. Schulz Co-founder, The Document Foundation, Kurfürstendamm 188, 10707 Berlin Gemeinnützige rechtsfähige Stiftung des bürgerlichen Rechts Legal details: http://www.documentfoundation.org/imprint Mobile Number: +33 (0)6 98 65 54 24. -- To unsubscribe e-mail to: [email protected] Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/ Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette List archive: http://listarchives.libreoffice.org/global/users/ All messages sent to this list will be publicly archived and cannot be deleted
