Activity for Harald Oehlmann
-
Harald Oehlmann posted a comment on ticket #338
Martin, I know the developper. I will ask them to perhaps contribute the code. Stay tuned. Take care, Harald Am 28.08.2025 um 19:01 schrieb Git Lost: Hi, yes it would be nice to support JAB Code and it is on the agenda at some stage. It having an open source encoder and decoder available already is huge, but the main problem is that the code is GPL licensed so we'd have to implement are own version to comply with our BSD-3- Clause licence. [tickets:#338] https://sourceforge.net/p/zint/tickets/338/...
-
I have tested the proposed list solution in advance., what is identical what is implemented: https://github.com/apnadkarni/twapi/commit/42bb48fee7110297090d1a10bb35321cd2169a59 This works and spaces in service names are supported by all other commands. THanks for all and take care, Harald
-
service.tcl: missing "list" in after command
-
-
Very important feature request!
-
Great, thanks , great solution !
-
GS1 linter: alpha checksum is 2 characters, not one
-
VIN Code prefix
-
What I observe in Industry is, that the people having installed this in the 90th are retired and nobody remembers any more the intention. I suppose, AIAG has a document and a decision on this, and ODETTE or JAIF may also, but those documents are out of sight of todays ingeniers equipped with an iPhone view of the world. Take care, Harald
-
Maybe, the following information is related: The "I" is the "Data Indentifier" for a VIN Code following ISO/IEC 15418. https://www.e-d-c.info/images/edc/documents/ANSI%20MH10.8.2_CM2020-08-27+.pdf Harald
-
Allow tcl demo to load zint from parent or neighbour foulder by default
-
Rebuild tcl windows build system
-
Merge branch 'master' of ssh://git.code.sf.net/p/zint/code
-
Add zint_tcl project files
-
Add nmake.vc build system, add tcl 9 compatibility, half-fix unload, half-add version-info
-
Sorry, round parenthesis are legal data, but not square brackets. Harald
-
Yes, they are. And they are not illegal. GS1 calles it the character set 82. That is, why zint uses "[10]A(10)b", as "[" is not within the character set 82, while "(" is included. So, the upper data will create a code with the correct lot number "A(10)b". Harald
-
Out of Spec PDF417 code with 18 columns and 52 rows
-
Yes, my fault. It is fixed with the current version. 2.9.1.9. was used.
-
Out of Spec PDF417 code with 18 columns and 52 rows
-
Daniel, the Swiss cross was the Swiss payment code. They tried to get it as CEN standard and to get this, the cross was removed. In addition, the copyright Owner of QR-Code (Denso Wave) constantly stops QR-code creators which are not normative, for example by including an image. Take care, Harald Am 28.06.2024 um 17:39 schrieb Daniel Gredler: An interesting use case that I ran across recently which blurs the line between "standard" and "non-standard" is the Swiss QR Code, which requires that a Swiss...
-
Here are the creators: https://www.qr-code-generator.com/about/ Take care, Harald
-
I have no idea who is behind that web site, sorry. At least, they have included the trademark for QR-Code (denso wave). The whois points to France: https://www.whois.com/whois/barcode-generator.org and points to Egoditor UG This was a German startup accquired now by a US company. In Germany, you are normally not allowed to have a web site without responsible person, data protection information etc. Quite curious, thank you. Nevertheless, here are the zint people. We like our own solution ;-). There...
-
I decided to use the patch with full capabilities. A border is just beautiful and it is helpful to find links. I looked into pdflatex and it used color "0 1 1" for hyperlinks and "0 0 1" for internal links. That looks good. So, here is the manual entry: hyperlinkAdd x y width height url ?options? x y width height is the box sensible for the hyperlink click url is the destination URL Possible options are: * -borderwidth: width of the border around the bounding box in display pixels, 0 to hide. Default:...
-
Dear Peter, please find attached the patch for the command: hyperlinkAdd x y width height url where a hyperlink clickable area is added to the given rectangular PDF region. I had a version which implements a lot of options like: hyperlinkAdd x y width height url -borderwidth 3 -borderradius 15 -bordercolor {0 0 255} -borderon 10 -borderoff 2 -borderhighlight I but this looks old style. tclfpdf just implements the upper command and I think, this is reasonable. Nevertheless, pdflatex shows borders...
-
Make distance of lineSpacing observable
-
-align right is not respected, if non-standard characters are replaced by "?"
-
Add font capabilities for more than 256 characters - use-case: Chinese Output
-
Add hyperlink capabilities
-
::pdf4tcl::createFontSpecEnc may check for maximum codepoint count
-
zint also has the possibility to not use auto-mode, but to specify the ECI. So, you may always use utf-8 if you like. Remark, that the ECI may also be changed in the middle of the message. This is helpful for further optimization or to mark parts of the message as binary data (ECI899) to avoid its transmission using keyboard interface. As you have remarked, there is room for improvements. I don't remeber the dynamics. It may also be that the functionality of the sister project BWIPP was synced into...
-
Hi Axel, thanks for the message. ZINT minimizes the encoded code size. That is why utf-8 is not used only. There was a standardisation attempt ( ISO/IEC7430 ), which was mutually killed by ISO/IEC SC31 and may be restarted after a change of project leader (less coleric person than myself). This document tried to set-up an ECI list which is guranteed to be implemented in decoders with utf-8 output and internal conversion. The list there is 3-13,15-18,20,26,28,30,32,899 China insists of GB18030 (ECI...
-
2.6 encodes the following codewords: 230 88 88 40 8 107 147 59 67 126 206 78 126 144 97 254 129 28 ECC: 122 222 250 19 198 18 11 153 105 34 3 128 34 93 And decodes to: "A1B2C3D4E5F6G7H8I9J0" So, it latches directly to C40 character set. 2.12 encodes the following codewords: 66 50 230 94 193 46 113 113 252 65 172 133 55 84 221 254 129 28 ECC: 80 81 50 147 211 47 29 48 245 96 91 225 133 217 And decodes to: "A1B2C3D4E5F6G7H8I9J0" So, 2.12 encodes "A1" in ASCII mode and then latches to C40. I don't understand,...
-
Extend stubs require string to an open string
-
Comply to TCL/Tk 9 C library. See https://core.tcl-lang.org/tcl/wiki?name=Migrating+C+extensions+to+Tcl+9
-
Due to that, I provided the gif encoder years ago. I use it to avoid the libpng dependency.
-
geturl not working with twapi::tls_socket anymore
-
Hi Simon, thanks for the message, great. Sad, that the project is abandoned. Your signature must be from one of the cross certificates still valid: https://learn.microsoft.com/en-us/windows-hardware/drivers/install/cross-certificates-for-kernel-mode-code-signing Before, the list was very long. Now, most certificates expired. Greetings from Naumburg (Saale), Harald
-
Dear Team, thank you for the great software ! I can confirm, that on my WIndows 10 64 bit box: The version in the download does not work The version in the post above by Simon Harston - 2021-09-16 works It would be great to have this version in the official distribution, so people don't stumble over the issue. I personally liked this tutorial about driver signing: https://www.davidegrayson.com/signing/ The issue is, that Microsoft has expired most cross certificates. My company has recently purchased...
-
Yes, I have read those pages. No hurry, no issue. Just for curiosity. It is very black magic. Thank you and take care, Harald
-
COM: control server parameter type
-
Make it easier to press Windows+u using great TWAPI
-
Hi Martin, I think, that is a good decission. It is the forward way. The incompatibility is limitted as passing non-ascii via CLI is always a mess. Great work ! Harald Am 10.05.2023 um 02:16 schrieb Git Lost: Hi Harald, As you can see above you were right about going for the wide interface on Windows. The fix [15fdca] https://sourceforge.net/p/zint/code/ci/15fdca2a0348de00892e7f3ebdf29f3ac301576a/ as it stands breaks backwards-compatibility for client API programs, by assuming that |outfile| and...
-
Hi Martin, thanks ! As always, I write unrelated stuff without digging in the whole context. I think, using the native codepage (and thus xxxA) for the file name list is just ok. This limits the choice, but this is IMHO ok. Command line etc. is to my knowledge quite fragile in Windows. Thanks, Harald
-
Hi Marcel, Martin, is it an option to use the wide interface on Windows ? CreateDirectoryA -> CreateDirectoryW ? Sorry, my comments are without code inspection etc. But nowdays, all the xxxA API commands are replaced by the xxxW version (since Windows 98, if I remember right). Harald
-
-
Thank you for the great discussion. To my knowledge, ECI and structure append is generally not allowed in GS1 mode. I am working on this to support ECI, at least ;-). The used App may be the free REA scan App. It is based on ZXING to my knowledge. ZXING is very weak in respect to all this.
-
Thank you for the great discussion. To my knowledge, ECI and structure append is generally not allowed in GS1 mode. I am working on this to support ECI, at least ;-). The used App is the free REA scan App. It is based on ZXING to my knowledge. ZXING is very weak in respect to all this.
-
I have made tests with ooxml. Anything I need works with the tclvfs::zip included in the current file (Version 1.0.4). So there is no need for action, and this is a "nice to have". Thank you for all, Harald
-
Update tcl::vfs to trunk in tclkit executable
-
FNC1 character is missing at beginning when creating a GS1 Barcode
-
I have checked it, it works correctly. You missed to check "GS1 mode" on the 2nd tab.
-
Did you check the "GS1 Option" on the QR-Code tab ?
-
svn-multi and spaces in file names
-
THank you. I suppose, this is fixed with bug #206. Ticket may be closed. Thank you, Harald
-
Thank you, TWAPI release 4.7.2 fixes this. Thanks for making a real release for this issue. Bug may be closed. THanks for all, Harald Just as a side not for people, who may repeat this: if a custom verify command is used, the error decission is on the command. So, with the following configuration, the issue is not detected: http::register https 443 {::twapi::tls_socket -verifier TWAPIVerifier} proc TWAPIVerifier {args} { return 1 } % set h [http::geturl https://expired.badssl.com] ::http::4 % http::status...
-
Thanks, Ashok, for sending the corrected version for the fconfigure -error issue. I repeat all tests and only mention the so far failed tests. fconfigure -error fconfigre -error reported a script error before, and not the real socket error. So, lets try it again: % set so [twapi::tls_socket -async expired.badssl.com 443] rc0 % catch {fconfigure $so -error} err derr 0 % set err Das empfangene Zertifikat ist abgelaufen. -> correct, error returned without error level. use case of http Within the http...
-
Preface Thank you for the enhanced version of tls.tcl. Please allow me to repeat the tests here for this use-case. iocp was active and it was renamed to socket. Sorry, I forgot this detail. package require iocp_inet rename ::socket ::socket_ori rename ::iocp::inet::socket ::socket In addition, twapi socket is registered to the http package. autoproxy ran, but probably did not do anything. fileevent Test, if the standard scripted fileevent still fires. The event queue did not run between socket and...
-
Preface Thank you for the enhanced version of tls.tcl. Please allow me to repeat the tests here for this use-case. fileevent Test, if the standard scripted fileevent still fires. The event queue did not run between socket and fileevent: % proc pargs {args} {puts "<[join $args ,]>"} % set so [twapi::tls_socket -async expired.badssl.com 443];fileevent $so writable [list pargs $so] <rc0> Test ok. fileevent after update In contrast to the former version, the fileevent now fires, even if the update ran....
-
Dear Ashok, sorry, I was not notified about your post. I forgot to subscribe. Here is the proposed test: % proc pargs {args} {puts "<[join $args ,]>"} % set so [twapi::tls_socket -async expired.badssl.com 443]; fileevent $so writable [list pargs $so] <rc0> Yes, in this case, it works. So, the issue is, that the update should not be run jet. Sorry, that was stupid and logic. The important role of "update" got only clear to me after some experimentations. Thanks also for the solution provided by private...
-
Preface Thank you for bringing this annoying issue up into a small example. This is probably due to the weakness of the stacked channel architecture in error cases. I suppose, the same issues apply to TCLTLS. It is probalbly a stacked channel issue and not a TWAPI issue. Nevertheless, Ashok is in a good position to look after this. If I can be of any help, please advice. I may retry to trigger Andreas to look into this regarding stacked channels. Expert time is a rare resource and we are all voluntears....
-
Preface Thank you for bringing this annoying issue up into a small example. This is probably due to the weakness of the stacked channel architecture in error cases. I suppose, the same issues apply to TCLTLS. It is probalbly a stacked channel issue and not a TWAPI issue. Nevertheless, Ashok is in a good position to look after this. If I can be of any help, please advice. I may retry to trigger Andreas to look into this regarding stacked channels. Expert time is a rare resource and we are all voluntears....
-
Thank you for the 4.7.1 starkits. This was delivered the day to the client. I will report here, if the issue arises again. Thank you so much, Harald
-
HTTP TLS background error "key "Socket" not known in dictionary"
-
Please add a top-level directory to the files in the tarball
-
Martin Gieseking, thanks, we will care, sorry for the inconvenience. The other Martin, will you care ? Give me a sign, if I should take care. Harald
-
I have scanned the sample code. It works great, thank you! Result, see attachment.
-
I have verified it. Just choose Code128 as symbology and it will work out. Take care, Harald
-
Add EFN format (German Medical Association)
-
And, I think, on the gui side, we may always use highest compression. The gui is never time and memory constraint. There, the option may only be used for testing. Thank you all, Harald
-
Thank you for the discussion. Yes, there are zint user for constraint systems, which care about calculation speed and specially about memory usage. Zint is used in a couple of embedded printing systems. So, it might be helpful, to take care about them. But I think, a more optimal bar code is the aim of zint and thus, the agressive method may be the default. We may think about a general swithc like "-lowmem/-lowspeed", where we may disable costly optimizations in any codes. Other usages may be to...
-
Thank you, Ashok, I appreciate. Take care, Harald
-
datamatrix size 144x144 barcode is not readable using the zxing online decoder
-
Hi André, Hi Martin, I can confirm that many decoders have issues with the short quiet zone of Data Matrix. QR-Code has 3 modules quiet zone. It is a great analysis that it only concerns 144x144. I propose to ask the question on the zxing side. Thank you, Harald
-
Dear Michael, thank you for your request. I am personally quite unhappy with those solutions, as QR-Code is not designed to be partly covered. Thus, putting a logo somewhere in a QR requires: an increase of the error correction level make sure, that no inner finder pattern is covered Zint will hopefully not support this, as a lot of know-how is required. As an alternate solution, you may use a symbology which is designed for that, for example JAB-Code: https://www.iso.org/standard/76478.html I can...
-
Great improvement, thank you!
-
Hi Gregory, thank you for the proposal. Unfortunately, iQR is not heavily supported. It is now superseeded by the DENSO solution rmQR. The corresponding ISO/IEC standard is in DIS phase and may already be purchased. AFAIK, iQR will not be pushed to get and ISO/IEC standard and will not be widely implemented. Here is the link to the document: https://www.iso.org/standard/77404.html Thank you, Harald
-
TWAPI documentation: command "cert_system_store_locations" does not have a LOCATION parameter
-
Thank you for the answer, I appreciate. I would prefer to change the man page (format) that all keywords may be marked and indexed. Thank you, Harald
-
"yield" not found in search index
-
I now did a full compilation from the win folder: anything works well ! Well, I had to remove the "[string totitle img::bmp]" in the pkgIndex.tcl from: package ifneeded img::bmp 1.4.13 [list load [file join $dir tkimgbmp1413t.dll] [string totitle img::bmp]] Otherwise, I get the error: % package require img::bmp cannot find symbol "Img::bmp_Init" So, the pkgIndex.tcl line, which works for bmp is: package ifneeded img::bmp 1.4.13 [list load [file join $dir tkimgbmp1413t.dll]]
-
base/win/makefile.vc: it is PRJ_PACKAGE_TCLNAME, not PRJ_PACKAGE_NAME
-
Consider "package require img::window" instead Img
-
2021-01-14 HaO TCL Backend: Added detection of presence of the Tk package and late initialization. This is a preparation to add a TCL only mode to the DLL.
-
Added changelog in TCL file, that character encoding is now treated by zint and TCL encodings are not used any more.
-
QR in EMF has border
-
Maybe, the border is added by Word/Paint ? Are you aware of the no background option ?
-
make qtzint on WIndows: Instructions updated.
-
2021-01-05 2.9.1 HaO TCL backend update
-
Thank you for all your efforts. I catch now the TWAPI calls. Normally, I am not to much interested in those errors. If a negociation fails, it is like there is nothing to do. It is just ignored. At maximum, it is logged. Thank you, Harald
-
Ashok, I have no clue. Now I have an application, where I always ignore the first call after the clipboard monitor, and now, the initial call is not coming any more. I have no clue why. I tried to set clibpard in advance or nat etc, but it does not fire any more, when the command is initiated. 5 minuites ago, it still did. Sorry, Harald
-
Control TLS socket server errors
-
buffer var not filled
-
Icon-only ttk::menubutton -style Toolbutton have large extra empty space on the right
-
pckIndex wrong for 9.2.2
-
Changes in README:
-
Document recent changes in README:
1 >