Wow 68ms that was a massive speed increase I didnt realise that line was a progress dialogue although I did wonder about how many times it did whatever it does when I asked you about n. I would never have expected it alone was the cause though.
You are way in front of me, I don't understand or know how to program or code but I can see that things are going in the right direction and as a layman, I am looking forward to trying the next version. For my part as an ordinary user I can send feedback as best as I can and I'm hoping that will be helpful in some way to progress the program to stability and usability. I haven't felt like this since I first discoved Shadow Defender in it's infancy. I love the idea of single person developers, breaking new ground, it cheers me up!
I am also impressed with Stefan's work on this and his enthusiasm, we talked about the idea and he had a test version ready faster than you could shake a stick at it
Yes it's a great thing RL. it lifts my spirits. I remember in the early days of the internet when I'd see people making little useful programs before all the later hype, boobytrapware, advertware, rentware, bloatware and phone homes. There is some great, really fantastic software today but I look back fondly to those times.
Stephan maybe you don't need to update the progress indicator on every iteration of the loop try it on every 100th iteration
Thanks RockLobster and sdmod ! I finally figured out why the binary file encoding/decoding did not work without base64. I had forgot to use unsigned char cast in the at() method so instead of giving the index range of the defaultPalette 0 - 255 it gave index range of -128 to 127 which covers just nicely the printable ASCII codes but nothing more. Also, the default.txt palette file only has 255 lines while it should have 256 and also it has one duplicate value too ....
v1.2 is out now! Sorry that took this long ... I was little distracted by my DNS project. Anyway, biggest changes now are: - Should not crash anymore (keeps fingers crossed ....) - PDF viewing support, but only for Linux users right now (will add for Windows users too if someone can give me library called poppler-qt5 crosscompiled for Windows and also all it's needed dependency libs too cross-compiled for Win... *sight*) - Added "Settings" tab where are things like compression level, encoding/decoding progress update interval, and option to decide if use base64 or not. - Finally, if decoding image that imagesecret can't open in viewer area it will now at least show some little information of the file (and saving of course is always possible) More complete description of all the stuff in ChangeLog.txt Links & screenshots: Code: https://www.orwell1984.today/ImageSecret_1.2_src.zip https://www.orwell1984.today/ImageSecret_1.2_XP.zip https://www.orwell1984.today/ImageSecret_1.2.zip http://ukp5un24mpxbqcpu.onion/ImageSecret_1.2_src.zip http://ukp5un24mpxbqcpu.onion/ImageSecret_1.2_XP.zip http://ukp5un24mpxbqcpu.onion/ImageSecret_1.2.zip PDF viewing on Linux. Note: To make it easier to view on small screens there is an splitter area between encoding and decoding section so to help hiding the encoding section. New "Settings" tab
Stefan, Thanks very much for your work and your communication. Things are going in th right direction I've just had the chance of getting five minutes to test. The XP version of IS 1.2 is much faster with the files (very useable) and has a logical GUI layout and sequence of actions. I tried it with a couple of fairly large pdfs (10.5 MB) and there were no crashes and I liked that the 'decode' and 'save file' buttons were greyed out until the file was opened to be decoded. The original files was re-constituted without error and/but needed re-naming, which is a plus (security/privacy) and a minus (identification) at the same time if you can't remember the name of the original file. I will test more files and larger later. Can you send me a PM to test a new file of yours to see if this XP version will read a file encoded by the main version. Your original link with palette file that you pmd me is dead. I'm still not quite sure what the Show Base64 tickbox function is for? Sorry if I have missed anything, I haven't had much time today. best wishes sdmod
Sure thing, I PM you new file to test. I'm still not quite sure what the Show Base64 tickbox function is for? Sorry if I have missed anything, I haven't had much time today. best wishes sdmod[/QUOTE] Oh, that show base64 thing is just if user is interested to know what the ascii presentation of the file looks like. I was not really sure if I should keep it now that base64 is optional and off by default. And I also found out that that there is trouble showing very large files (30 MB +) base64 output (it just shows nothing) so maybe I will remove the whole "show base64" tickbox in next release.
Thanks Stefan, I received your pm with palette file and . png link and all worked with no problem and decoded fine.. I just used the xp version 1.2 making random palette on 31.2 MB .wav file resulting .png image was 93.4 MB Took slightly longer but no problem and the file played as normal when decrypted. then I tried to encode and decode 47.5 MB .pdf file which didn't work....the end file came out as a 0KB file, although it had showed correct size prior to finish MIME Type: application/pdf Comment: PDF document Preferred suffix: pdf Size: 49902728 bytes 'Show Base64' tickbox was unticked on both files, successful and un-successful sdmod Oh, that show base64 thing is just if user is interested to know what the ascii presentation of the file looks like. I was not really sure if I should keep it now that base64 is optional and off by default. And I also found out that that there is trouble showing very large files (30 MB +) base64 output (it just shows nothing) so maybe I will remove the whole "show base64" tickbox in next release.[/QUOTE]
Stefan, I didn't keep the large png that I mentioned in the post but I've tried it again using the same pdf and although the finished file wasn't a 0 it was 32.0 MB this time it still didn't open the finished file properly. The problem for me sending it to you, It's quite large to upload. Wilders file upload in PM didn't seem to want to accept it. If you try it with a large pdf of about the same size you will probably get the same result as me. Let me know how you get on. My original pdf was 47.5 MB (49,902,728 bytes) Could it be anything to do with processing time for a file that size? I've no idea, just a thought. The outcome file for the 30mb ish one came out about three times more size. best wishes sdmod I will have more time later today to do more tests.
Hmmm...I gave it 30.9 MB pdf file and the output is 84.5 MB png file so it's about 2.7 times (about three times like you said) bigger. But we have to take into account that pdf files are binary files so they don't compress as well as text files. Without compression my 30.9 MB would be around 123 MB. Each input byte has to be encoded as 32-bit pixel output, that's 4 bytes per pixel. So without any compression any file would produce image the size of x4 input. I have been thinking...do your machine has enough memory? I have not checked how much the compression and encoding process takes memory. My test machine has 16 GB. EDIT: You are right, it has problems with large files. I feeded it this time pdf the size of 52 MB and output png only 1.55 MB Have to check whats wrong.... EDIT2: Got little idea how to, maybe, cut the file overhead to half or so but have to make some test code first....
Stefan, I'm glad that it wasn't just me Although I have much less ram my system copes pretty well with most things 32 bit Xp and 4gb ram although XP can't use it all. Please keep us updated on how you get on and if you have anything to test. sdmod
Yeah I was wondering what happened when we didn't hear from Stefan. I hope mapping the internet for his portable DNS server didn't get him into some kind of trouble.
Heh ... not at all It's just that my vacation has started and still trying to recover from jet lag ... So development of IS and also portable DNS will continue but lil slower now.
ImageSecret 1.3 source code now available at: Code: https://www.orwell1984.today/imagesecret.html This version fixes few bugs (especially the very bad bug introduced by previous version, the breaking of encoding/decoding when doing alpha invisiblity!) Also in this version: dummy pixel support! (right now, only fixed color but maybe already in next version will have the ability to randomize it) Premade binaries (for 2 windows version and for linux) come later when I have time. (If you are in a hurry send me a bribe and I prioritize )
Thats looking good Stefan I wish I could test it myself but like I said in pm there are some issues preventing me doing that.
Hi Stefan, I've PMd you. I would appreciate it if you would let me know when you have a version that I can use. Patrick