Re: 80 million downloads
On 12/3/13 10:28 PM, David Gerard wrote: On 3 December 2013 21:12, jan i j...@apache.org wrote: I have (for ubuntu) tried to read through the policies which was a nightmare. The first thing would be to get into Debian. I recall the steps there were set out (basically, have an install set that doesn't conflict in any way with LO - however fair or unfair that might be, it's what you need to do). MATE managed it in the face of GNOME 3, so AOO should be able to. I have no problem to make this more public. If LO want to be independent they should not use our package names, binary names etc. Juergen - d. - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: Good progress on confusing advertisements on download page
On 12/4/13 1:04 AM, Roberto Galoppini wrote: 2013/12/3 Roberto Galoppini roberto.galopp...@gmail.com: 2013/12/3 FR web forum ooofo...@free.fr: A question for us is: How do we want to report misleading ads in the future? Do we need any process on our side? Do we want to track the reports in Bugzilla? Or do we want to just report these as individuals, possibly leading to redundant reports? Maybe record these in BZ (where adding image attachments and de-duping is easy) and then report to SF with a link to the BZ issue? Good question I try to download AOO in french and SF.net display this kind of ad. http://oooforum.free.fr/images/sf_ad.png A link to a fake version named OpenOffice 2013 This jump to a para-site: h**p://opf.derniereversion.info Thanks for heads up on this, I'll take care of that. It has been reviewed and we asked to remove that, it might take a business day or two (worst case). thanks Roberto, I am sure SF will benefit from this as well. It won't be good for SF if users get the wrong stuff and are not satisfied by the offered service. Juergen - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: Improvements of OUString
On 12/4/13 7:20 AM, Andrew Douglas Pitonyak wrote: On 12/03/2013 08:32 AM, Herbert Duerr wrote: On 03.12.2013 13:02, Andre Fischer wrote: Additionally to this almost correct statement one could mention that isEmpty() is preferred over getLength()0 and why. Yes, it is preferred for checking the emptiness because it directly expresses what it checks. In general it is a good idea to check for emptiness instead of counting the elements and then comparing against zero. Its the old interface vs. implementation detail question. The result will be the same from a mathematical standpoint but the effort to get this result may be different. From an algorithmic complexity standpoint an emptiness check is always equal or better. Maybe a mathematician can provide some insights from the set theory on this question? By the way: the String class of Java=6 got its isEmpty() method for the same reasons. As Herbert said, this directly expresses what it checks. I would add and it reads better. There are other reasons, however; for example, depending on the implementation, it may be significantly faster than comparing against the length. For one implementation it may not matter, but, if you change the implementation in the future, it may. Uniformity is another possible reason (assuming other classes also implement isEmpty()). If you become used to using the expressive form (ie, using isEmpty() rather than comparing size), then you can leave optimizing the call to the class implementer rather than worrying about the correct way to do it. I have seen empty strings checked by comparing the length to zero, and even using a string compare against an empty string. and again to replace getLength() with isEmpty() can be something for newbies to get more familiar with the code and do something useful. Juergen - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: 80 million downloads
From: Jürgen Schmidt [mailto:jogischm...@gmail.com] I have no problem to make this more public. If LO want to be independent they should not use our package names, binary names etc. in fact, so you're absolutely right. I am a normal Linux user and the situation on Linux is extremely confusing. But what can we do? I think here are problems that need to be resolved legally. Greetings, Jörg - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: Improvements of OUString
On 04.12.2013 07:20, Andrew Douglas Pitonyak wrote: As Herbert said, this directly expresses what it checks. I would add and it reads better. There are other reasons, however; for example, depending on the implementation, it may be significantly faster than comparing against the length. For one implementation it may not matter, but, if you change the implementation in the future, it may. Yes. Especially considering that the length of a string is somewhat ambiguous. Currently getLength() returns the number of codepoints in the strings encoding (e.g. UTF-16). But one could also reasonably define the length of a string as the number of unicode codepoints. This number is different from getLength() if unicode surrogates were involved or unicode normal forms were taken into consideration. Uniformity is another possible reason (assuming other classes also implement isEmpty()). If you become used to using the expressive form (ie, using isEmpty() rather than comparing size), then you can leave optimizing the call to the class implementer rather than worrying about the correct way to do it. For historic reasons the STL and TR1 containers use empty() to check for emptiness and clear() to empty the container. In the previous sentence I intentionally used empty once as an adjective and once as a verb to show that the method name empty() is ambiguous. Therefore I preferred to use the Java inspired method name isEmpty() for checking the emptiness of a string. I have seen empty strings checked by comparing the length to zero, and even using a string compare against an empty string. Indeed: In our current codebase there are all kinds of variants for this simple check. E.g. aString == OUString() aString.equalsAscii() aString.equals( OUString()) aString.compareTo( OUString()) aString.compareToAscii( ) aString.getLength() == 0 aString.getLength() 1 and the related checks: aString.getLength() 0 aString.getLength() = 1 Also there was the now already eliminated buggy variant if( aString) With the new method all these could be cleanly consolidated. I must admit that the long discussion thread over the seemingly simple new method isEmpty() is somewhat scary to me... Herbert - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Patches on Windows (Update)
I have just checked in what basically is a reimplementation of the defunct support for creating patches on and for Windows. Building OpenOffice without the new release=t flag should work like before. If you don't plan to build patches then you can stop reading now (just tell me if your builds break). Support for patch creation is still in its early stages. It may or may not work. Having said that, I would not mind if you would try it out, just don't apply them yet outside a virtual machine. At the moment I am interested to get the workflow right and reliably working. It is a lengthy process, something in the area of 30 minutes per language, and I can make only so many test runs. But you don't have to run it to point out how to improve the workflow. Here is how I currently think how patch creation can work. Note that patch creation is something that I expect is only done once per release by the release manager (or whoever builds the release installation sets.) It may sound very technical but it really list only the most important points. 0. Expected (disk) space and time requirements: Disk space: Number of languages * - Previous downloadable installation set (EXE) in ext_sources/ - The above unpacked in instsetoo_native/wntmsci12.pro - The CAB file of the above (maybe 95% of the installation set) unpacked - The installation set and downloadable installation set of the current release - A copy of the above, arranged so that msimsp.exe can process it. - The MSP patch, about 10% of the downloadable installation set. I did not measure it but I would estimate the total to be around 500 MB Time: - Download previous installation set from archive.apache.org (use a local cache if you can) - Unpack the above - Unpacke the CAB inside the above - Build the installation set of the current version - Make a local copy of the above - Run msimsp.exe All in all about 30 minutes 1. Prepare the release builds (one per language) that are compatible with the previous release. Why: There are a lot of conditions to fulfill before you can create a patch between two installation sets. One is that files must not be deleted. Another is that files keep the same internal data (unique name, sequence number, etc.) These conditions have to be checked and enforced. How: Go to main/instsetoo_native and run 'build release=t' This flag should probably be added as option to configure. It is not used outside of instsetoo_native but it is important that it is not forgotten when the installation sets are built. What: In order to check the conditions, solenv/bin/make_installer.pl (which is called by build/dmake from instsetoo_native) needs access to the previous release. It does that by reading the necessary data from the previous installation sets. For that a) the downloadable installation set is downloaded from archive.apache.org If you are concerned about too many downloads from archive.apache.org: - Creating patches is typically done twice a year, by the release manager - You can provide a local repository. This also speeds up this step. b) the downloadable installation is unpacked to get access to its content, most importantly (in this stage) the MSI file. For this I use 7zip, the only program I know of, that is able to unpack the installation set EXE that is created by the NSIS tool. c) use the msidb.exe program from the Windows SDK to read the database tables in the MSI 2. Store information about the new release for when we will be creating patches when doing the next (future) release. Why: The current release will eventually be the previous release. How: The tool is not checked in, I am still cleaning it up a bit. What: This extends the main/instsetoo_native/data/releases.xml list with data about the current release. 3. Create the patches (one per language) How: Go to main/instsetoo_native/util and run 'dmake patch_create' What: a) Check if the installation sets for the relevant languages are present (both the previous and current ones) b) Check that all conditions (that I know of) are fulfilled. Patch creation will break when even one condition is not met. It will, for example, detect when the release builds where not made with the release=t flag. c) Create a local copy of the current release. This is necessary because the msimsp.exe tool (the one that creates the actual patch file) needs the CAB file (the container that contains all files that are to be installed) and its unpacked content in the same directory. d) Run the msimsp.exe tool to create the patch file. Creating an installation set, unpacking and copying the previous and current ones, are already lengthy processes that each can take several minutes to complete. Msimsp.exe usually takes 10 to 15 minutes to run. 4. (For testing) Apply the patch Why: To verify that the created patch really works Attention: While steps 1 to 3, if
Re: EXTERNAL: Re: Building comphelper
Hi Raymond, Thanks for the information. Sorry about the PDF. Unfortunately, we are unable to do a simple copy and paste because of the way are system is setup. The only other option that I have is to type the output, but that would have taken much longer. Ok. I just wanted to make sure that its not a simple oversight. C++ template errors are often very extensive and look quite scary. If the problem hides in a little type difference of the n-th argument of a multi-layer template specialization, then finding these miniscule differences is hard enough even with regexp magic or with specialized tools (template error message decryptors) such as STLFilt. Without the error message text but just images of this text it becomes much harder to diagnose a problem. We just implemented the patch, but are still receiving the same output. We are investigating now. This cannot possibly be! The original error output was that construction of the insertion object failed because of type and constness mismatches. With the patch there is no insertion object anymore, so there error output cannot possibly be the same! Herbert - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Checked out revision 1547453 Build problem on Windows 7
Hello,
Windows 7, Cygwin64 Terminal
With command:
$ svn co https://svn.apache.org/repos/asf/openoffice/trunk aoo-trunk
Receive AOO upgrade Checked out revision 1547453.
Try to rebuild it with command:
$ build --all
build -- version: 275224
Have error messages in sal module:
=
Building module sal
=
Entering /cygdrive/c/source/aoo-trunk/main/sal/inc
mkdir.exe -p ../wntmsci12.pro/slo/pch
precompiled_sal.cxx
mkdir.exe -p ../wntmsci12.pro/slo/pch_ex
precompiled_sal.cxx
Entering /cygdrive/c/source/aoo-trunk/main/sal/osl/all
Compiling: sal/osl/all/utility.cxx
Compiling: sal/osl/all/debugbase.cxx
C:/PROGRA~2/MICROS~1.0/VC/include\xhash(60) : error C2039: 'basic_string' :
is n ot a member of '_STL'
C:/PROGRA~2/MICROS~1.0/VC/include\xhash(60) : error C4430: missing type
specifie r - int assumed. Note: C++ does not support default-int
C:/PROGRA~2/MICROS~1.0/VC/include\xhash(60) : error C2143: syntax error :
missin g ',' before ''
C:/PROGRA~2/MICROS~1.0/VC/include\xhash(887) : error C2143: syntax error :
missi ng ';' before ''
C:/PROGRA~2/MICROS~1.0/VC/include\xhash(887) : error C2059: syntax error :
''
C:/PROGRA~2/MICROS~1.0/VC/include\xhash(887) : error C2065: '_Traits' :
undeclar ed identifier
C:/PROGRA~2/MICROS~1.0/VC/include\xhash(888) : error C2143: syntax error :
missi ng ';' before '{'
C:/PROGRA~2/MICROS~1.0/VC/include\xhash(888) : error C2447: '{' : missing
functi on header (old-style formal list?)
It seems a problem with MS VS include module:
C:\...oft Visual Studio 9.0\VC\includexhash
The last successful build was with AOO Checked out revision 1537066.
Could you please help me to understand what is wrong now?
Currently I'm not subscribed to the Development Mailing List, please put me
in CC in your Reply .
Thank you,
Vadim.
Re: Improvements of OUString
On 04.12.2013 11:12, Herbert Duerr wrote: On 04.12.2013 07:20, Andrew Douglas Pitonyak wrote: As Herbert said, this directly expresses what it checks. I would add and it reads better. There are other reasons, however; for example, depending on the implementation, it may be significantly faster than comparing against the length. For one implementation it may not matter, but, if you change the implementation in the future, it may. Yes. Especially considering that the length of a string is somewhat ambiguous. Currently getLength() returns the number of codepoints in the strings encoding (e.g. UTF-16). But one could also reasonably define the length of a string as the number of unicode codepoints. This number is different from getLength() if unicode surrogates were involved or unicode normal forms were taken into consideration. Uniformity is another possible reason (assuming other classes also implement isEmpty()). If you become used to using the expressive form (ie, using isEmpty() rather than comparing size), then you can leave optimizing the call to the class implementer rather than worrying about the correct way to do it. For historic reasons the STL and TR1 containers use empty() to check for emptiness and clear() to empty the container. In the previous sentence I intentionally used empty once as an adjective and once as a verb to show that the method name empty() is ambiguous. Therefore I preferred to use the Java inspired method name isEmpty() for checking the emptiness of a string. I have seen empty strings checked by comparing the length to zero, and even using a string compare against an empty string. Indeed: In our current codebase there are all kinds of variants for this simple check. E.g. aString == OUString() aString.equalsAscii() aString.equals( OUString()) aString.compareTo( OUString()) aString.compareToAscii( ) aString.getLength() == 0 aString.getLength() 1 and the related checks: aString.getLength() 0 aString.getLength() = 1 Also there was the now already eliminated buggy variant if( aString) With the new method all these could be cleanly consolidated. I must admit that the long discussion thread over the seemingly simple new method isEmpty() is somewhat scary to me... Seemingly is the operative word. Personally, I find aString.getLength() 1 much more scary. Do you want to extend the documentation of OUString with some of what we found out in this thread (getLength(), isEmpty(), etc) or do you want somebody else to do it? -Andre Herbert - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: EXTERNAL: Re: Building comphelper
Hi Raymond, Just to give you an update. We were able compile debugbase.cxx by including /opt/solarisstudios12.3/prod/include/CC/stlport4 Ah, please don't. Stlport4 is dead. , but the next module wanted ../include/CC/Cstd. Then it went back and forth. There seems to be a disconnect between which C/C++ implementation it wants and I do not think they are compatible with each other. Yes, they are not compatible to each other. As of now, we've loaded the Apache C++ stdcxx4 implementation so that we can retry using just the one implementation. However, we've noticed that stdcxx4 does not provide hash_map and hash_set, among others, Indeed: hash_map and hash_set were replaced by unordered_map and unordered_set in the C++ standardization process. In our codebase we'll have to replace them but for now they are already mapped to their newer counterparts using wrappers in the main/stlport/systemstl/hash_* headers. Please make sure that you are using --without-stlport in your configure option so that this wrapping becomes active. Herbert - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: Good progress on confusing advertisements on download page
2013/12/4 Jürgen Schmidt jogischm...@gmail.com: On 12/4/13 1:04 AM, Roberto Galoppini wrote: 2013/12/3 Roberto Galoppini roberto.galopp...@gmail.com: 2013/12/3 FR web forum ooofo...@free.fr: A question for us is: How do we want to report misleading ads in the future? Do we need any process on our side? Do we want to track the reports in Bugzilla? Or do we want to just report these as individuals, possibly leading to redundant reports? Maybe record these in BZ (where adding image attachments and de-duping is easy) and then report to SF with a link to the BZ issue? Good question I try to download AOO in french and SF.net display this kind of ad. http://oooforum.free.fr/images/sf_ad.png A link to a fake version named OpenOffice 2013 This jump to a para-site: h**p://opf.derniereversion.info Thanks for heads up on this, I'll take care of that. It has been reviewed and we asked to remove that, it might take a business day or two (worst case). thanks Roberto, I am sure SF will benefit from this as well. It won't be good for SF if users get the wrong stuff and are not satisfied by the offered service. That's right Jürgen, and let me add all ads under our direct control have never displayed confusing or misleading ads. Unfortunately this is not true for ads delivered by advertising network partners, that serving millions of ads have lesser control on what is displayed. It's an on-going process, and community help is highly appreciated. Roberto Juergen - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Subscribe me
Reporting a problem with the OpenOffice website
Hi there I have discovered that people receiving an attachment in Open Office, who are using Microsoft Office 2007 cannot open my attachment. It indicates the attachment is corrupted. Is there any way around this ? Thanks. Ian Ogilvie ian.ogilvie...@gmail.com
Re: Reporting a problem with the OpenOffice website
On Wed, 4 Dec 2013 12:46:38 + Ian Ogilvie ian.ogilvie...@gmail.com wrote: Hi there I have discovered that people receiving an attachment in Open Office, who are using Microsoft Office 2007 cannot open my attachment. It indicates the attachment is corrupted. Is there any way around this ? Thanks. Ian Ogilvie ian.ogilvie...@gmail.com What format was transmitted? -- Rory O'Farrell ofarr...@iol.ie - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
RE: EXTERNAL: Re: Building comphelper
Thanks Herbert. Do you have any idea why we would be receiving the following? Compiling: sal/rtl/source/unload.cxx /usr/local/include/boost/unordered/detail/emplace_args.hpp, line 199: Error:Could not find a match for boost::tuples::getboost::tuples::N,boost::tuples::HT, boost::tuples::TT(constboost::tuples::tupleboost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type) needed inboost::unordered::detail::construct_from_tupleconfigmgr::Partial::Node,boost::tuples::null_type(configmgr::Partial::Node*, constboost::tuples::tupleboost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type). /usr/local/include/boost/unordered/detail/emplace_args.hpp, line 350:Where: While instantiatingboost::unordered::detail::construct_from_tupleconfigmgr::Partial::Node,boost::tuples::null_type(configmgr::Partial::Node*, constboost::tuples::tupleboost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type, boost::tuples::null_type,boost::tuples::null_type, boost::tuples::null_type). /usr/local/include/boost/unordered/detail/emplace_args.hpp, line 350:Where: Instantiated from boost::unordered::unordered_maprtl::OUString,configmgr::Partial::Node, boost::hashrtl::OUString,_STL::equal_tortl::OUString, _STL::allocator_STL::pairconst rtl::OUString,configmgr::Partial::Node::operator[](const rtl::OUString). /opt/aoo-4.0.0/main/ sal/rtl/source/unload.cxx , line 217: Where: Instantiated from non-template code. -Original Message- From: Herbert Duerr [mailto:h...@apache.org] Sent: Wednesday, December 04, 2013 3:54 AM To: dev@openoffice.apache.org Cc: Steele, Raymond; Meffe, David K Subject: Re: EXTERNAL: Re: Building comphelper Hi Raymond, Just to give you an update. We were able compile debugbase.cxx by including /opt/solarisstudios12.3/prod/include/CC/stlport4 Ah, please don't. Stlport4 is dead. , but the next module wanted ../include/CC/Cstd. Then it went back and forth. There seems to be a disconnect between which C/C++ implementation it wants and I do not think they are compatible with each other. Yes, they are not compatible to each other. As of now, we've loaded the Apache C++ stdcxx4 implementation so that we can retry using just the one implementation. However, we've noticed that stdcxx4 does not provide hash_map and hash_set, among others, Indeed: hash_map and hash_set were replaced by unordered_map and unordered_set in the C++ standardization process. In our codebase we'll have to replace them but for now they are already mapped to their newer counterparts using wrappers in the main/stlport/systemstl/hash_* headers. Please make sure that you are using --without-stlport in your configure option so that this wrapping becomes active. Herbert - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Printing a database
Hi I have created a database of my DVD collection. The creation of it was fine, but there does not seem to be an option to print. If it can be done, can you tell me how. I am running Windows 8 Regards David McGhee - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Open Office ECCN query
Good afternoon, I am working on a CGI IT UK Limited project called 'Galileo' on which we are looking to export software. As part of this process, we need to know whether Open Office is of United States origin, and if so what the ECCN number is, and does an ENC licence also apply. I would be most grateful if you are able to provide me with this information. However, failing this could you direct me to where I might find the information I require? Your assistance with this matter is most appreciated. Best regards, Richard Richard Trebilcock | ILS Engineer Defence | CGI Chaucer House, Springfield Drive, The Office Park, Leatherhead, Surrey, KT22 7LP UK T: +44 01372 838258 | M: +44 (0) 7717 355882 richard.trebilc...@cgi.com | cgi-group.co.uk [cid:image001.jpg@01CE2A2B.EE6E9D30] CGI IT UK Limited. A CGI Group Inc. Company Registered Office: 250 Brook Drive, Green Park, Reading RG2 6UA, United Kingdom. Registered in England Wales - Number 947968 CONFIDENTIALITY NOTICE: Proprietary/Confidential Information belonging to CGI Group Inc. and its affiliates may be contained in this message. If you are not a recipient indicated or intended in this message (or responsible for delivery of this message to such person), or you think for any reason that this message may have been addressed to you in error, you may not use or copy or deliver this message to anyone else. In such case, you should destroy this message and are asked to notify the sender by reply e-mail.
Re: 80 million downloads
On Wed, Dec 4, 2013 at 2:02 AM, Jörg Schmidt joe...@j-m-schmidt.de wrote: From: Jürgen Schmidt [mailto:jogischm...@gmail.com] I have no problem to make this more public. If LO want to be independent they should not use our package names, binary names etc. in fact, so you're absolutely right. I am a normal Linux user and the situation on Linux is extremely confusing. But what can we do? I think here are problems that need to be resolved legally. Greetings, Jörg - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org One of the confusing issues comes from LO's continued use of soffice as the launchpad binary. We've been down this road before to no avail. See, for example -- http://www.libreoffice.org/developers/ On SOME distros who handle this a little better/politely, the LO binary has been renamed to libreoffice, and a symlink from soffice to libreoffice (/usr/bin/libreoffice --a custom packaging ) is supplied which can easily be removed. bottom line -- even if we are successful in getting AOO into some repositories, how to deal with this naming clash. Maybe someone who has downloaded LO packages directly from their site can provide more information. -- - MzK Cats do not have to be shown how to have a good time, for they are unfailing ingenious in that respect. -- James Mason
Re: 80 million downloads
On 04-Dec-2013, at 13:19, Kay Schenk kay.sch...@gmail.com wrote: On Wed, Dec 4, 2013 at 2:02 AM, Jörg Schmidt joe...@j-m-schmidt.de wrote: From: Jürgen Schmidt [mailto:jogischm...@gmail.com] I have no problem to make this more public. If LO want to be independent they should not use our package names, binary names etc. in fact, so you're absolutely right. I am a normal Linux user and the situation on Linux is extremely confusing. But what can we do? I think here are problems that need to be resolved legally. Greetings, Jörg - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org One of the confusing issues comes from LO's continued use of soffice as the launchpad binary. We've been down this road before to no avail. See, for example -- http://www.libreoffice.org/developers/ On SOME distros who handle this a little better/politely, the LO binary has been renamed to libreoffice, and a symlink from soffice to libreoffice (/usr/bin/libreoffice --a custom packaging ) is supplied which can easily be removed. bottom line -- even if we are successful in getting AOO into some repositories, how to deal with this naming clash. Maybe someone who has downloaded LO packages directly from their site can provide more information. When was the last time we actually contacted one of the people at LO who can effect a change or at least diminish the confusion? I mean, it's silly to have the legacy identity and suggests either a rather tired play at the real soffice or laziness. Yes, we could change our package name but…. not only do we have a large installed base but also legacy, copyright, and actual growth. - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
draft blog post: Call for Comments: Apache OpenOffice Distributor Best Practices
https://blogs.apache.org/preview/OOo/?previewEntry=call_for_comments_apache_openoffice This idea has been in my queue for a while. I'd like to push it out now for public comment, and based on that try to get a new distributor page set up by January. -Rob - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: Submission for consultants page
On Tue, Oct 22, 2013 at 6:33 PM, Rob Weir robw...@apache.org wrote: On Tue, Oct 22, 2013 at 4:07 PM, Drew Jensen drewjensen.in...@gmail.com wrote: Howdy Rob, Ah trademarks - yes, I agree the Apache OpenOffice should have a TM, a la: IBM® Lotus® Symphony™ is a suite of open source office applications. I've put in a request to get that page updated. Just to close this out, I can see that the page has now been updated: http://www-03.ibm.com/software/products/en/ibm-support-for-apache-openoffice Regards, -Rob A couple of quick questions if I may. Will the support only be available to those with the IBM connectors? IIRC that was the plan in the past. It is certainly possible to combine IBM Support for Apache OpenOffice along with a license for IBM Connections or SmartCloud. In that case our Connector [1] would be supported as well. We also have an integration solution for OpenOffice and IBM Enterprise Content Management [2]. We're exploring other integrations as well. That said, if a customer wants help with a large, complex, standalone deployment of Apache OpenOffice, we'll talk with them. That was also the case with Symphony. Also, does IBM have any plans to list AOO support on the United States GSA registry? This was available in the past but not available for the last few years. I have no idea. Should we? Regards, -Rob [1] http://extensions.openoffice.org/en/project/ibm-connections-connector-apache-openoffice [2] http://www-01.ibm.com/software/data/information-agenda/catalog/profiles/OpenOffice_Integration_IBM_ECM.html Thanks, //drew On Mon, Oct 21, 2013 at 7:44 PM, Jörg Schmidt joe...@j-m-schmidt.de wrote: From: Rob Weir [mailto:robw...@apache.org] Sorry, but I think the website does not comply with the necessary formalities. Always has been emphasized, these are marked on the relevant websites trademark of Apache, for example: IBM Support for Apache OpenOffice TM and not only: IBM Support for Apache OpenOffice I think that would be more confusing. Sorry, but appearance is imho not so much the issue here, but rather trademark rights of Apache. Yes, that is formal, but how else should this be handled? It would suggest the entire name was a trademark since it is a continuous name with capital letters. It does not have to be in the headline. Put identification in continuous text, for example in: IBM® Support for Apache OpenOffice™ offers expert technical support for Apache OpenOffice™, ... where IBM is already marked with ®. Greetings, Jörg - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: 80 million downloads
On 4 December 2013 20:10, Andrew Rist andrew.r...@oracle.com wrote: On 12/4/2013 10:19 AM, Kay Schenk wrote: On Wed, Dec 4, 2013 at 2:02 AM, Jörg Schmidt joe...@j-m-schmidt.de wrote: From: Jürgen Schmidt [mailto:jogischm...@gmail.com] I have no problem to make this more public. If LO want to be independent they should not use our package names, binary names etc. in fact, so you're absolutely right. I am a normal Linux user and the situation on Linux is extremely confusing. But what can we do? I think here are problems that need to be resolved legally. Greetings, Jörg - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org One of the confusing issues comes from LO's continued use of soffice as the launchpad binary. We've been down this road before to no avail. See, for example -- http://www.libreoffice.org/developers/ On SOME distros who handle this a little better/politely, the LO binary has been renamed to libreoffice, and a symlink from soffice to libreoffice (/usr/bin/libreoffice --a custom packaging ) is supplied which can easily be removed. I believe we should create a plan to have a binary named openoffice with a symlink to soffice. +1 we need to protect the copy right. The longer we allow the distros to use openoffice, the less likely it is that they will (and need to) change. rgds jan I. We should announce this very publicly and communicate to LO and the distros when we make the change. This would also be a good topic to discuss with LO at FOSDEM this year. It would be good to get agreement from LO that they agree with this approach and will (over time) implement this across all of the distros (as it is a reasonable and fair way to deal with the overlap between the products) A. bottom line -- even if we are successful in getting AOO into some repositories, how to deal with this naming clash. Maybe someone who has downloaded LO packages directly from their site can provide more information. - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: draft blog post: Call for Comments: Apache OpenOffice Distributor Best Practices
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Rob Weir wrote: https://blogs.apache.org/preview/OOo/?previewEntry=call_for_comments_apache_openoffice This idea has been in my queue for a while. I'd like to push it out now for public comment, and based on that try to get a new distributor page set up by January. -Rob Rob; Change is the leader free and... to is the leading free and... Change CDs to CD's. In the last paragraph; Change wide Apavh Community to wider Apache Community. Other than that it looks good. -BEGIN PGP SIGNATURE- Version: GnuPG v2.0.22 (MingW32) Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/ iQEcBAEBAgAGBQJSn40+AAoJEH0fu5UhGmBCdQEIAKZI9G5JX3BBHPkIq3Qsh7eF G96bsatFtIgjmTy5hR2k5N5G1mHiPkvFSzaFjuNCWZWPDDHXhbfUHVYjdxWFn2Vd F1PwfNGXGKFCw1CehjGN6eMOAKfp2pYvnNdtEdrET/AJSOoVNa+h5RFZr8RyVttS UoO/Q8TTrb2TizzmfqPI2K2d3afBvG9RtxB+dDr5VBfCtTiapHxbo6MQBUCzT+ab C38tU/ObsufBE1NZGnwO4BqP0QagtzH1B9l/vWjBfGaRAnUGK2sDcAtUMQ2T9NPE wb+oLapp5NRBA0HNJZQXE9PL11EVKuHHc19ueLP9a0bRZdx69VHpI2fd++uqh4U= =UAoo -END PGP SIGNATURE- - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: draft blog post: Call for Comments: Apache OpenOffice Distributor Best Practices
On Wed, Dec 4, 2013 at 3:14 PM, Keith N. McKenna keith.mcke...@comcast.netwrote: Change CDs to CD's. Why?
RE: soltools need(s) to be rebuilt
Hello, I am still struggling... Now I was trying to build libxmlsec and the configuration parameters did not give the expected result. I had to enter manually the following command ./configure ADDCFLAGS= CPPFLAGS= --with-pic --disable-shared --disable-crypto-dl --with-libxslt=no --with-openssl=no --with-gnutls=no LIBXML2LIB=-lxml2 --enable-pkgconfig=no --with-openssl=/usr --with-nss=/extra/sources/OpenOffice/aoo4/main/solver/410/unxsogi.pro --with-nspr=/extra/sources/OpenOffice/aoo4/main/solver/410/unxsogi.pro in order to build the libraries. Is this OK? There is a libcrypto.a in /extra/sources/OpenOffice/aoo4/main/solver/410/unxsogi.pro/lib but this cannot be used to build a shared object (at least that's what the linkers say). Regards, A.S. -- Apostols Syropoulos Xanthi, Greece
Re: 80 million downloads
Kay Schenk wrote: On SOME distros who handle this a little better/politely, the LO binary has been renamed to libreoffice, and a symlink from soffice to libreoffice (/usr/bin/libreoffice --a custom packaging ) is supplied which can easily be removed. Actually, the official LibreOffice packages install to their own subtree of /opt by default, so they do not install conflicting aliases (but those packages do still use program/soffice). Packages that come with distributions, instead, usually store all binaries in /usr/bin and in that case we have the double soffice conflict. I think unopkg may have the same problem. This naming is chosen by the individual distributions, even though dropping soffice completely may have some (possibly limited) side effects. For more details, see the relevant discussions at: https://lists.fedoraproject.org/pipermail/devel/2013-February/178173.html http://nabble.documentfoundation.org/No-usr-bin-soffice-symlink-on-Linux-td4034854.html Regards, Andrea. - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: draft blog post: Call for Comments: Apache OpenOffice Distributor Best Practices
On Wed, Dec 4, 2013 at 3:27 PM, Donald Whytock dwhyt...@apache.org wrote: On Wed, Dec 4, 2013 at 3:14 PM, Keith N. McKenna keith.mcke...@comcast.netwrote: Change CDs to CD's. Why? OK. I've always wondered about this myself, so I looked it up. It looks like there are various guidelines, but generally the aim is to avoid confusing the user. One style guideline (New York Times) says to use the apostrophe in cases where an acronym uses periods, but not otherwise. So M.D.'s, but CDs. http://afterdeadline.blogs.nytimes.com/2010/04/13/faqs-on-style/?_r=0 -Rob - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: draft blog post: Call for Comments: Apache OpenOffice Distributor Best Practices
On Wed, 4 Dec 2013 15:49:43 -0500 Rob Weir robw...@apache.org wrote: On Wed, Dec 4, 2013 at 3:27 PM, Donald Whytock dwhyt...@apache.org wrote: On Wed, Dec 4, 2013 at 3:14 PM, Keith N. McKenna keith.mcke...@comcast.netwrote: Change CDs to CD's. Why? OK. I've always wondered about this myself, so I looked it up. It looks like there are various guidelines, but generally the aim is to avoid confusing the user. One style guideline (New York Times) says to use the apostrophe in cases where an acronym uses periods, but not otherwise. So M.D.'s, but CDs. http://afterdeadline.blogs.nytimes.com/2010/04/13/faqs-on-style/?_r=0 -Rob Look up Greengrocer's apostrope. The ' indicates possive case, or abbreviation of is or has (but not in this case) John's trousers (= the trousers of John) John's gone home (= John has gone home) -- Rory O'Farrell ofarr...@iol.ie - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: draft blog post: Call for Comments: Apache OpenOffice Distributor Best Practices
On Wed, 4 Dec 2013 22:05:08 + Rory O'Farrell ofarr...@iol.ie wrote: On Wed, 4 Dec 2013 15:49:43 -0500 Rob Weir robw...@apache.org wrote: On Wed, Dec 4, 2013 at 3:27 PM, Donald Whytock dwhyt...@apache.org wrote: On Wed, Dec 4, 2013 at 3:14 PM, Keith N. McKenna keith.mcke...@comcast.netwrote: Change CDs to CD's. Why? OK. I've always wondered about this myself, so I looked it up. It looks like there are various guidelines, but generally the aim is to avoid confusing the user. One style guideline (New York Times) says to use the apostrophe in cases where an acronym uses periods, but not otherwise. So M.D.'s, but CDs. http://afterdeadline.blogs.nytimes.com/2010/04/13/faqs-on-style/?_r=0 -Rob Look up Greengrocer's apostrope. The ' indicates possive case, or abbreviation of is or has (but not in this case) Should read 'possessive' I was being heckled by Real Life in the middle of correcting a sticky 's' key and hit Send too soon. John's trousers (= the trousers of John) John's gone home (= John has gone home) -- Rory O'Farrell ofarr...@iol.ie - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: Arrow heads with hole
Hi Juergen, yes, of course we should add that stuff, that's what I had in mind. I already have a grep on the task, just wanted to give some background information. thanks go to regina for alwas bringing something forward! -- ALG (iPad) Am 04.12.2013 um 09:07 schrieb Jürgen Schmidt jogischm...@gmail.com: On 12/3/13 6:05 PM, Armin Le Grand wrote: Hi Regina, On 02.12.2013 23:45, Regina Henschel wrote: Hi Andrea, Andrea Pescetti schrieb: On 26/11/2013 Regina Henschel wrote: I have expanded the standard.soe with some arrow heads with hole. The file is attached to https://issues.apache.org/ooo/show_bug.cgi?id=123758. If you like them, we can consider to use this palette as default. I added a screenshot to the issue for clearer comparison. The new styles are nice and it would be good to have them in 4.1. In some cases, in the preview, I see the main line of the arrow going (seemingly) too far within the arrow head, see http://imagebin.org/280257 for an example. Is this wanted? No, that is not wanted. I will explain the problem in more detail: First: nice arrowheads! I am currently on something else, but I have already a grep on the task which you wrote. Background: - The arrows traditionally only used polygons (topologically: no poly-polygons which allows holes). This is alrteady changed and implemented. - The arrow heads and line overlapping traditionally use a 0.0 or 0.8 factor of overlap, according to 0% or 80% overlap. These values are handles relative to the arrow head sizes already, but are treated as a boolean (one or the other). In the core these values are prepared to be 0-100%, freely choosable. In the UI you can switch between 0% and 80% in the lines dialog in it's first page (line ends- 'centered' for right and left arrow). Missing is to have a value input instead of that simple switch and to get that 0-100% value over the API and to ODF. This is where normally Regina knows if this is possible ;-) Thus: Old. historical limitations, some more way to go to get over these. When one day it will be possible to choose that value freely (prepared in core and primitives) you will be able to trim these to connect to your arrow head as you want it to be. If there would be a way to do this automatically could also be considered; the old overlapping paint was probably only implemented since no one wanted to do it better from the beginning (time constrains? other?). Regina, this sounds as if we could use a feature task for his... Do I understand correct that with Reginas fix the arrows looks much better and the problem is less obvious. If yes I would still integrate it because it is an improvement. And if possible improve it further later on. But we should not wait for a 100% solution that most of the user even don't recognize. Well that is only my personal opinion. Juergen If you stop the line at the very place where the arrow head starts, you get a visible gap between the square 45° and the line itself for fat lines (and same for circle or any peak shape). Therefore an overlap was introduced. For the filled arrow heads, it does not matter whether the line is drawn a little bit longer. For the arrow heads with hole you have to find a compromise between showing a gap at the outer part and showing a little bit line in the hole. Currently the amount by which the line is drawn longer does not depend on the kind of arrow head, but on the length of the arrow head. It is in file polygonprimitive2d.cxx in method PolygonStrokeArrowPrimitive2D::create2DDecomposition around line#547 the statement fStart *= 0.8; In LO I have changed that to fStartOverlap = getStart().getWidth() / 15.0;, so that it depends on the width of the arrow head, which also determines the 'stroke' width in the non-filled arrow heads. It is a compromise too. (It is not really a 'stroke', but the area between two combined paths.) We could copy that in AOO. But perhaps someone has a better idea? Kind regards Regina - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org
Re: draft blog post: Call for Comments: Apache OpenOffice Distributor Best Practices
On Wed, Dec 4, 2013 at 2:42 PM, Rory O'Farrell ofarr...@iol.ie wrote: On Wed, 4 Dec 2013 22:05:08 + Rory O'Farrell ofarr...@iol.ie wrote: On Wed, 4 Dec 2013 15:49:43 -0500 Rob Weir robw...@apache.org wrote: On Wed, Dec 4, 2013 at 3:27 PM, Donald Whytock dwhyt...@apache.org wrote: On Wed, Dec 4, 2013 at 3:14 PM, Keith N. McKenna keith.mcke...@comcast.netwrote: Change CDs to CD's. Why? OK. I've always wondered about this myself, so I looked it up. It looks like there are various guidelines, but generally the aim is to avoid confusing the user. One style guideline (New York Times) says to use the apostrophe in cases where an acronym uses periods, but not otherwise. So M.D.'s, but CDs. http://afterdeadline.blogs.nytimes.com/2010/04/13/faqs-on-style/?_r=0 -Rob It looks fine and I think some of users would appreciate such a listing with the assumption that we would at least do some sort of vetting of the CD vendor/contents. Look up Greengrocer's apostrope. The ' indicates possive case, or abbreviation of is or has (but not in this case) Should read 'possessive' I was being heckled by Real Life in the middle of correcting a sticky 's' key and hit Send too soon. John's trousers (= the trousers of John) John's gone home (= John has gone home) -- Rory O'Farrell ofarr...@iol.ie - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org -- - MzK Cats do not have to be shown how to have a good time, for they are unfailing ingenious in that respect. -- James Mason
Re: Printing a database
On Wed, Dec 4, 2013 at 9:34 AM, dmcghee m...@dmcghee.plus.com wrote: Hi I have created a database of my DVD collection. The creation of it was fine, but there does not seem to be an option to print. If it can be done, can you tell me how. I am running Windows 8 Regards David McGhee - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org Hello David, If you want to print the entire contents of a database table, it is easiest to import the table into Calc or Writer. If you registered your database when you created it, the process is simple. 1. Open a new Calc or Writer document. 2. Press F4, which will open a pane near the top of the screen showing all of your registered data sources. 3. Click the + next to the name of your database to show the tables and queries. 4. Expand the list of tables using the +. 5. Click and drag the name of your table into the new document. 6. Print that document. I hope that helps. Best Regards, Francis
Re: Printing a database
On Wed, Dec 4, 2013 at 9:34 AM, dmcghee m...@dmcghee.plus.com wrote: Hi I have created a database of my DVD collection. The creation of it was fine, but there does not seem to be an option to print. If it can be done, can you tell me how. I am running Windows 8 Regards David McGhee - To unsubscribe, e-mail: dev-unsubscr...@openoffice.apache.org For additional commands, e-mail: dev-h...@openoffice.apache.org Another copy in case David isn't registered. Hello David, If you want to print the entire contents of a database table, it is easiest to import the table into Calc or Writer. If you registered your database when you created it, the process is simple. 1. Open a new Calc or Writer document. 2. Press F4, which will open a pane near the top of the screen showing all of your registered data sources. 3. Click the + next to the name of your database to show the tables and queries. 4. Expand the list of tables using the +. 5. Click and drag the name of your table into the new document. 6. Print that document. I hope that helps. Best Regards, Francis
Re: Arrow heads with hole
On 12/5/13 12:34 AM, Armin Le Grand wrote: Hi Juergen, yes, of course we should add that stuff, that's what I had in mind. I already have a grep on the task, just wanted to give some background information. thanks go to regina for alwas bringing something forward! well I don't understand why you grep the task, I believe Regina is knowing quite well what she is doing. We want to grow our developer community and want to motivate people to work on the code on their own. We all know that Regina is already working independent and she need if at all mainly some interlock to discuss, brainstorm certain issues, problems or solutions. Independent of this I believe that the experienced developers should give more guidance to others and help to solve their problems. Even if it takes longer. This is the way to move forward, let people work on the code and with the code and provide guidance where necessary. We don't need to discuss this further and I hope I made my point clearer. And I hope Regina simply integrate here good work ;-) Thanks Regina -- ALG (iPad) Am 04.12.2013 um 09:07 schrieb Jürgen Schmidt jogischm...@gmail.com: On 12/3/13 6:05 PM, Armin Le Grand wrote: Hi Regina, On 02.12.2013 23:45, Regina Henschel wrote: Hi Andrea, Andrea Pescetti schrieb: On 26/11/2013 Regina Henschel wrote: I have expanded the standard.soe with some arrow heads with hole. The file is attached to https://issues.apache.org/ooo/show_bug.cgi?id=123758. If you like them, we can consider to use this palette as default. I added a screenshot to the issue for clearer comparison. The new styles are nice and it would be good to have them in 4.1. In some cases, in the preview, I see the main line of the arrow going (seemingly) too far within the arrow head, see http://imagebin.org/280257 for an example. Is this wanted? No, that is not wanted. I will explain the problem in more detail: First: nice arrowheads! I am currently on something else, but I have already a grep on the task which you wrote. Background: - The arrows traditionally only used polygons (topologically: no poly-polygons which allows holes). This is alrteady changed and implemented. - The arrow heads and line overlapping traditionally use a 0.0 or 0.8 factor of overlap, according to 0% or 80% overlap. These values are handles relative to the arrow head sizes already, but are treated as a boolean (one or the other). In the core these values are prepared to be 0-100%, freely choosable. In the UI you can switch between 0% and 80% in the lines dialog in it's first page (line ends- 'centered' for right and left arrow). Missing is to have a value input instead of that simple switch and to get that 0-100% value over the API and to ODF. This is where normally Regina knows if this is possible ;-) Thus: Old. historical limitations, some more way to go to get over these. When one day it will be possible to choose that value freely (prepared in core and primitives) you will be able to trim these to connect to your arrow head as you want it to be. If there would be a way to do this automatically could also be considered; the old overlapping paint was probably only implemented since no one wanted to do it better from the beginning (time constrains? other?). Regina, this sounds as if we could use a feature task for his... Do I understand correct that with Reginas fix the arrows looks much better and the problem is less obvious. If yes I would still integrate it because it is an improvement. And if possible improve it further later on. But we should not wait for a 100% solution that most of the user even don't recognize. Well that is only my personal opinion. Juergen If you stop the line at the very place where the arrow head starts, you get a visible gap between the square 45° and the line itself for fat lines (and same for circle or any peak shape). Therefore an overlap was introduced. For the filled arrow heads, it does not matter whether the line is drawn a little bit longer. For the arrow heads with hole you have to find a compromise between showing a gap at the outer part and showing a little bit line in the hole. Currently the amount by which the line is drawn longer does not depend on the kind of arrow head, but on the length of the arrow head. It is in file polygonprimitive2d.cxx in method PolygonStrokeArrowPrimitive2D::create2DDecomposition around line#547 the statement fStart *= 0.8; In LO I have changed that to fStartOverlap = getStart().getWidth() / 15.0;, so that it depends on the width of the arrow head, which also determines the 'stroke' width in the non-filled arrow heads. It is a compromise too. (It is not really a 'stroke', but the area between two combined paths.) We could copy that in AOO. But perhaps someone has a better idea? Kind regards Regina - To unsubscribe, e-mail:
