{"payload":{"feedbackUrl":"https://github.com/orgs/community/discussions/53140","repo":{"id":682308998,"defaultBranch":"main","name":"shellChatGPT","ownerLogin":"mountaineerbr","currentUserCanPush":false,"isFork":false,"isEmpty":false,"createdAt":"2023-08-23T22:36:48.000Z","ownerAvatar":"https://avatars.githubusercontent.com/u/92537363?v=4","public":true,"private":false,"isOrgOwned":false},"refInfo":{"name":"","listCacheKey":"v0:1692830209.402465","currentOid":""},"activityList":{"items":[{"before":"81ef89b171e91e838b3a970dab17d2684a0ab291","after":"82ed43bda4b0b58bc6d853e58da61d6186e530ee","ref":"refs/heads/main","pushedAt":"2024-09-17T20:19:29.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"sync","shortMessageHtmlLink":"sync"}},{"before":"8c938968efbca484e217db4403c517e377bb2582","after":"81ef89b171e91e838b3a970dab17d2684a0ab291","ref":"refs/heads/main","pushedAt":"2024-09-13T10:05:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"+ add o1-preview and o1-mini support.\n\nI havent got acccess to these models yet, but they are tentatively implemented (hopefully works!).\n\n---","shortMessageHtmlLink":"+ add o1-preview and o1-mini support."}},{"before":"bfb1313020f2debb337019068911e8fe94ef69a3","after":"8c938968efbca484e217db4403c517e377bb2582","ref":"refs/heads/main","pushedAt":"2024-09-13T09:48:31.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Add o1-preview and o1-mini support.\n\nI havent got acccess to these models yet, but they are tentatively implemented (hopefully works!).\n\n---","shortMessageHtmlLink":"Add o1-preview and o1-mini support."}},{"before":"3cdfd8a9c3ceb9838d09964dc97816bf0082f736","after":"bfb1313020f2debb337019068911e8fe94ef69a3","ref":"refs/heads/main","pushedAt":"2024-09-11T19:23:13.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Bug fixes and improvements.\n\nImprove last update on session_mainf() to trim spaces from both ends of input rather than only the trailing spaces.\n\nAdjust silence detection values in read_charrecf() for voice recording on whisper mode.\n\nUpdate docs.\n\n---","shortMessageHtmlLink":"Bug fixes and improvements."}},{"before":"aafdc530027fffd7f103923be4589e91ede82608","after":"3cdfd8a9c3ceb9838d09964dc97816bf0082f736","ref":"refs/heads/main","pushedAt":"2024-09-10T05:08:42.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Set Termux pulseaudio sles auto set to 4 secs.\n\n---","shortMessageHtmlLink":"Set Termux pulseaudio sles auto set to 4 secs."}},{"before":"074b96b0320af4d10a8c89c3d878e3172450260a","after":"aafdc530027fffd7f103923be4589e91ede82608","ref":"refs/heads/main","pushedAt":"2024-09-10T03:38:04.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Bug fixes and improvements.\n\nFix the function session_globf() when the argument is \"current\", and \"current\" points to a custom filepath for the chat history file.\n\nRemove trailing spaces from the argument to session_mainf().\n\nFix setting INSTRUCTIONS in a new chat when using chat command \"/break\". Instructions were not being backed up correctly.\n\n---","shortMessageHtmlLink":"Bug fixes and improvements."}},{"before":"2b2ea8f51803264c160068332e23c07c37fccbe6","after":"074b96b0320af4d10a8c89c3d878e3172450260a","ref":"refs/heads/main","pushedAt":"2024-09-03T12:26:13.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Fix Termux auto-set pulseaudio \"module-sles-source\".  Defaults to NO.\n\nWe revert defaulting to NO, do not set pa \"sles source\" automatically.\n\nWe keep the warnings as this is probably a setting the Termux user wants to add to his pulseaudio configuration.\n\nOnly print the warnings when options -vv are set, so we dont disrupt the old Termux way ofrecording andplaying which works just fine...\n\nThe pulseaudio with sles module is only required for the silence-detection feature, so lets not set it unless needed!\n\n---","shortMessageHtmlLink":"Fix Termux auto-set pulseaudio \"module-sles-source\". Defaults to NO."}},{"before":"d4557d85e8699bebc343dcec26ebaa7674dde3d4","after":"2b2ea8f51803264c160068332e23c07c37fccbe6","ref":"refs/heads/main","pushedAt":"2024-09-02T09:40:35.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Adjust down sleep time in set_termuxpulsef().\n\n---","shortMessageHtmlLink":"Adjust down sleep time in set_termuxpulsef()."}},{"before":"1595c71febaee7ac20b30c84063e6443941c6e02","after":"d4557d85e8699bebc343dcec26ebaa7674dde3d4","ref":"refs/heads/main","pushedAt":"2024-09-02T01:35:03.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Small optimisation of last code update.\n\n---","shortMessageHtmlLink":"Small optimisation of last code update."}},{"before":"8b7db11e97167f073770591eb4bd327f81b90ad5","after":"1595c71febaee7ac20b30c84063e6443941c6e02","ref":"refs/heads/main","pushedAt":"2024-09-02T01:16:07.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Termux voice record fixes and improvements.\n\nUpdate setting Termux pulseaudio \"sles source\" automaticallly to be able to play and record audio. Add a little pause after enabling sles source so the device can be turned on completely.\n\nDiscovered a problem with Whisper auto-context when using option -C (continue) in chat mode, which is now fixed.\n\nLower silence-detection sensitivity to catch when user stops speaking in medium-noisy environments in the handless-mode (options -ccwz -vv).\n\n---","shortMessageHtmlLink":"Termux voice record fixes and improvements."}},{"before":"5f0bd0ebef128ba5509999cd0a84212b7fc6680e","after":"8b7db11e97167f073770591eb4bd327f81b90ad5","ref":"refs/heads/main","pushedAt":"2024-09-01T18:45:22.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"+ Doc updates\n\n---","shortMessageHtmlLink":"+ Doc updates"}},{"before":"d7efaa32528fa8fbe0a9c0753312c4a70ca4fc4b","after":"5f0bd0ebef128ba5509999cd0a84212b7fc6680e","ref":"refs/heads/main","pushedAt":"2024-09-01T18:06:53.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"+ Update docs.\n\n---","shortMessageHtmlLink":"+ Update docs."}},{"before":"aa8b8f9e338f2d02c4f58c6097f58dff29699b34","after":"d7efaa32528fa8fbe0a9c0753312c4a70ca4fc4b","ref":"refs/heads/main","pushedAt":"2024-09-01T18:02:20.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Update docs.\n\n---","shortMessageHtmlLink":"Update docs."}},{"before":"a701e7f560dab889e9afac4acf4b12b67c76a93d","after":"aa8b8f9e338f2d02c4f58c6097f58dff29699b34","ref":"refs/heads/main","pushedAt":"2024-09-01T11:27:35.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Fix case statement glob order in _model_costf().\n\n---","shortMessageHtmlLink":"Fix case statement glob order in _model_costf()."}},{"before":"980bf24de4f1a401eddaa2f8103ce12b4b2623e4","after":"a701e7f560dab889e9afac4acf4b12b67c76a93d","ref":"refs/heads/main","pushedAt":"2024-09-01T11:05:53.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Bug fixes in the \"dump text/pdf/doc/url\" feature and in the text editor interface.\n\nThe somewhat recent code for \"dump text/pdf/doc/url\" feature needed some adjustments\nto work well when the dump returned empty OR with an exit code.\n\nIt was necessary to check for the error code outside the scope of exit codes\nof the user confimation prompt options, and outside of the range of\nexit codes  our script itself uses for internal signaling.\n\nThe script internal exit codes range from 180 to 220, or a little more\nnarrowly, between 190 to 210.\n\nImprove the text editor interface in how the \"-xx\" chat command or\npressing \"x\" in user prompts would\nreturn to the text editor with the correct INPUT.\n\nAdd new OpenAI model chatgpt-4o to known list of models, so that model capacity\nhas the right setting.\n\nUpdate known model pricing.\n\nRenamed the configuration variable $COST_CUSTOM to $MOD_PRICE, as teh previous name for this variable is really bad...\n\nAdded the shortcut to exit the script with pressing \"Q\" (uppercase) in the\nuser confirmation prompts (only when option -v is not set).\n\n---","shortMessageHtmlLink":"Bug fixes in the \"dump text/pdf/doc/url\" feature and in the text edit…"}},{"before":"f76cf193fa023e14a91bdbbec4d6a972af5fd955","after":"980bf24de4f1a401eddaa2f8103ce12b4b2623e4","ref":"refs/heads/main","pushedAt":"2024-08-30T02:49:27.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Bug fixes + update default models for Groq and Google Gemini services.\n\nUpdate default models\n\nGROQ: llama-3.1-70b-versatile\n\nGEMINI: gemini-1.5-pro-latest\n\n---\n\nCode cleaning. We clean some old code which must not affect how it works.\nMostly minor cleaning changes.\n\n---\n\nFix very important bug that would really only show up when running the script with options \"-vv\",\nin _sysmsgf() (old __sysmsgf) because it would return with 1 in case $OPTV was not less than 2.\n\nThis should be nothing but we use \"! _sysmsgf\" sometimes to return from other functions with\nan error code, but it would just flop the returning value to truw when it was false (when\noptions -vv were set).\n\n---\n\nThe function naming convention of the script was a little awkward and\nmany function names has double underscores before the function name.\nWe reduced the use of double underscores in name function to make it\nmore readble.\n\nThe confusion derives from what is now the _sysmsgf() function which was one of the first\nfunctions to be written and was kind of a base for the following function namings.\n\nAlso, chat mode would increase +1 $OPTV value, and _sysmsgf() heeds it,\nbut automatically increasing $OPTV in chat mode proved to be a bad idea\nand was reverted some time there after.\n\nHowever, the logic underlying _sysmsgf() was flawed at the beggining and it has changed\nso it has been time to change all functions that have weird names.\n\nBasically, one underscore means it is a helper function *or* the raw function\nof a feature, and two underscores means a helper of a helper function.\n\nNot perfect logic but hopefully it is now aligned.\n\n---\n\nThe The `+=' assignment operator is now used less frequently in the code to append strings to strings.\nWe only use this operator to append items to arrays and in arithmetic ops.\n\nWhy?\n\nAs this bash feature is so little used, and there is a more inter shell compatible ways of performing the same operations, and\nalbeit _that is not_ the goal to be strictly bourne-shell compatible, we prefer to decrease the use of bashisms as much as possible.\n\nbash-3.1-alpha1:\n\ni.  The `+=' assignment operator (append to the value of a string or array) is\n    now supported for assignment statements and arguments to builtin commands\n    that accept assignment statements.\n\n---","shortMessageHtmlLink":"Bug fixes + update default models for Groq and Google Gemini services."}},{"before":"364ed47c15f3cf783b3a2a6c41b99cd3050cf3e0","after":"f76cf193fa023e14a91bdbbec4d6a972af5fd955","ref":"refs/heads/main","pushedAt":"2024-08-29T01:26:48.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Comment out executing trim_silencef() function to remove silence from voice record.\n\nThe threshold is probably too low to actually remove any significant amount of \"silence\" from the voice record.\n\nThis works well with professionally recorded audio to remove silence periods, but voice record in random environments usually have a lot of background noise.\n\nSo I am commenting this function out as it is probably an overhead at this time. I want to test it more to see if it actually makes any significant \"silence\" removal.\n\n---","shortMessageHtmlLink":"Comment out executing trim_silencef() function to remove silence from…"}},{"before":"3a3e8ee52a3a83ec55be07dc915b71be3f2144ee","after":"364ed47c15f3cf783b3a2a6c41b99cd3050cf3e0","ref":"refs/heads/main","pushedAt":"2024-08-28T21:05:25.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Update is_mdf() to be more strict when checking for markdown in response.\n\nAs this function is used to auto-detect markdown in response and enable markdown rendering, it turns out that some answers only have ordered or unordered lists, and if that is the *only* markdown, it is not worth it turning on markdown automatically. However, is the lists are idented, then markdown will be turned on.\n\nMaybe in the future we will remove the markdown lists regex, because really just turn on markdown rendering when tehre is enough markdown to render!\n\n---","shortMessageHtmlLink":"Update is_mdf() to be more strict when checking for markdown in respo…"}},{"before":"611822de60ee34c5bbb7d86df26d0d87b4536375","after":"3a3e8ee52a3a83ec55be07dc915b71be3f2144ee","ref":"refs/heads/main","pushedAt":"2024-08-28T19:44:45.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Test and reduce the filetypes _is_docf() function can detect.\n\nIt turns out LibreOffice does not have so many options to convert-from from the command-line. I thought it would convert PPT / XLS to TXT for example, but it cannot. But it can handle DOC, DOCX, ODT and  RTF, which is good enough. Also, add abiword as it can handle these file types as well.\n\nfunction _is_docf\n{\n\ttypeset -l ext=$1\n\tcase \"$ext\" in\n\t*.doc|*.docx|*.odt|*.ott|*.rtf) :;;\n\t*) false;;\n\tesac;\n}\nfunction is_docf { \t[[ -f $1 ]] && _is_docf \"$1\" ;}\n\n---","shortMessageHtmlLink":"Test and reduce the filetypes _is_docf() function can detect."}},{"before":"29d9f337a14a42900d5210f3d9708a8553a12abb","after":"611822de60ee34c5bbb7d86df26d0d87b4536375","ref":"refs/heads/main","pushedAt":"2024-08-28T15:06:22.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Small documentation fixes.\n\n---","shortMessageHtmlLink":"Small documentation fixes."}},{"before":"9e39e2e43ad325a80e8403d201391beccbde4b3d","after":"29d9f337a14a42900d5210f3d9708a8553a12abb","ref":"refs/heads/main","pushedAt":"2024-08-28T14:51:56.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Vision / multimodal models change of input handling. Silently process filenames appended to the end of user prompt.\n\nBefore, it was required to separate filenames with a pipe \"|\" when a filename had white spaces.\nNow, pipe _is not_ used anymore as an operator-separator to delimit multimodal input.\n\nHowever, filenames with white spaces are now required to have backslash-escaped white spaces in the middle of the filepath!\n\nThis is the way shell scripts must work as it resonates with other shell-way of doing things.\n\nImprove how chat commands are executed when they invoke the \"/shell\" command internally.\n\nTurning on `$PSKIP` is very powerful in how it changes the procedure to take in the chat turn.\nIt must be strictly controlled, that is why now we have many added many safety measures to\nunset `$PSKIP` after it it serves it purpose.\n\nThis was not so much of an evident problem before but now that we execute the \"/shell\" command\na lot for automatic features (text dumps, mostly), we must take heed of `$PSKIP` closely.\n\nThis improves general user interface for redo / edit functions of the current user prompt.\n\nImprove the text editor interface a little, specially when user selects prompt \"redo\".\n\nImprove on the last few code updates.\n\nImprove the \"/pick\" to allow for picking filenames with white spaces.\n\n---","shortMessageHtmlLink":"Vision / multimodal models change of input handling. Silently process…"}},{"before":"e8285b7ca50ea1c1ac98c4685c570e82cffa5521","after":"9e39e2e43ad325a80e8403d201391beccbde4b3d","ref":"refs/heads/main","pushedAt":"2024-08-27T17:26:05.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Update code that dumps and appends text from supported file types at invocation.\n\n---","shortMessageHtmlLink":"Update code that dumps and appends text from supported file types at …"}},{"before":"86de928fe8916391da1c97e3195060090ae209f3","after":"e8285b7ca50ea1c1ac98c4685c570e82cffa5521","ref":"refs/heads/main","pushedAt":"2024-08-27T13:42:33.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Minor fix in last update....\n\n---","shortMessageHtmlLink":"Minor fix in last update...."}},{"before":"73a98acd75dceef566bfbc938162886377ad9837","after":"86de928fe8916391da1c97e3195060090ae209f3","ref":"refs/heads/main","pushedAt":"2024-08-27T13:39:58.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Decrease verbose messages when input for multimodal is invalid.\n\nThis improves the auto detection of what is likely not multimodal input. Automatic.\n\nAlso, try setting some function variables to empty when calling set_histf(), as this may (or may not) speed up the loop processing in this function...\n\nDecrease the number of file types _is_docf() can check, it was an noverhead to have all those document extensions. Also, add pdf type. _is_docf() has lower precedence than _is_pdf() function, which should work well together.\n\nAdd auto-completion option --openai to reset defaults to OpenAI in case command line invocation goes such as \"chatgpt.sh -cc --anthropic  --openai\", in which case anthropic option is unset.\n\n---","shortMessageHtmlLink":"Decrease verbose messages when input for multimodal is invalid."}},{"before":"a226bfa10537c494ee046b92cab7a8615e3e7d03","after":"73a98acd75dceef566bfbc938162886377ad9837","ref":"refs/heads/main","pushedAt":"2024-08-26T23:14:23.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Fix bug that would loop through the output of a shell command or file dump.\n\nFor example, running \"/sh wc ~/.bashrc\" would give \"wc\" and \"cat ~/.bashrc\" outputs! That is becaus ethe output of wc has the filepath, and that filepath would trigger to dump the bashrc text.\n\n---","shortMessageHtmlLink":"Fix bug that would loop through the output of a shell command or file…"}},{"before":"56a19ff88eff7298363a0fffce8812921a056a9e","after":"a226bfa10537c494ee046b92cab7a8615e3e7d03","ref":"refs/heads/main","pushedAt":"2024-08-26T21:50:00.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Minor code cleaning and straightening.\n\n---","shortMessageHtmlLink":"Minor code cleaning and straightening."}},{"before":"6614b58c9e12718bd5ce27d8646e9f8327db3885","after":"56a19ff88eff7298363a0fffce8812921a056a9e","ref":"refs/heads/main","pushedAt":"2024-08-26T19:14:29.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Fix and improve the code of the text/pdf/doc/url dump feature.\n\nWill now fail elegantly if libreoffice is not found, for example, as was expected before.\n\nAlso, the code for this feature needed some work and now is more readable because it uses less the parameter $var.\n\n---","shortMessageHtmlLink":"Fix and improve the code of the text/pdf/doc/url dump feature."}},{"before":"057b0819c8d0c93d4959124f2065549285c64453","after":"6614b58c9e12718bd5ce27d8646e9f8327db3885","ref":"refs/heads/main","pushedAt":"2024-08-26T15:39:30.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Update docs.\n\n---","shortMessageHtmlLink":"Update docs."}},{"before":"93b37a120f2b98ae28050eaa8f63aea3e7ed1b30","after":"057b0819c8d0c93d4959124f2065549285c64453","ref":"refs/heads/main","pushedAt":"2024-08-26T15:29:44.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Add support to easily dump text from .doc, .xls, etc file types.\n\nImportant: fix the request block json that affected a few specific situations when using the Anthropic API and its integrational code.\n\nUse LibreOffice when it is availiable to convert lots of different file types to text, such as word documents and spreadsheets.\n\nImproved how the automatic commands for pdf and doc dumps disable themselves in case no pdf-to-text or libreoffice suites are not found in the ssystem. This avoids the overhead of checking for the existence of the required software beforehand.\n\nAdd some comments in the source code with stuff that I am currently studying and peerhaps implement them in the code in the future, if needed.\n\nAdd sanity newlines.\n\nFix detecting the chat \"/cat command\" when there is arguents to it (optimisation regression).\n\nMinor adjustments of some varible substitution indexes and constants.\n\nMuch improved handling of files with spaces in paths for all the text dump functions (/cat, /pdf, and /doc).\n\nNew is_docf() function, and upgraded is_txturl() test function.\n\n```\nfunction _is_docf\n{\n\ttypeset -l ext=$1\n\tcase \"$ext\" in\n\t*.doc|*.docx|*.docm|*.dot|*.dotx|*.dotm|*.odt|*.ott|*.oth|*.rtf|*.uot|*.uof|\n\t*.xls|*.xlsx|*.xlsm|*.xlt|*.xltx|*.xltm|*.ods|*.ots|*.fods|*.uos|\n\t*.ppt|*.pptx|*.pptm|*.pot|*.potx|*.potm|*.pps|*.ppsx|*.ppsm|*.odp|*.otp|\n\t*.dif|*.slk|*.dbf|*.odg|*.otg|*.sxd|*.std|*.odf) :;;\n\t*) false;;\n\tesac;\n}\n\n```\n\n---","shortMessageHtmlLink":"Add support to easily dump text from .doc, .xls, etc file types."}},{"before":"05d5eb2655a4ac11883eb790f940fd4bdc18c4e9","after":"93b37a120f2b98ae28050eaa8f63aea3e7ed1b30","ref":"refs/heads/main","pushedAt":"2024-08-24T13:14:52.000Z","pushType":"push","commitsCount":1,"pusher":{"login":"bionota","name":"JSN","path":"/bionota","primaryAvatarUrl":"https://avatars.githubusercontent.com/u/84939133?s=80&v=4"},"commit":{"message":"Block _mediachatf() loop over 128 iterations (security).\n\nAlso, add some sanity newlines in _mediachatf().\n\n---","shortMessageHtmlLink":"Block _mediachatf() loop over 128 iterations (security)."}}],"hasNextPage":true,"hasPreviousPage":false,"activityType":"all","actor":null,"timePeriod":"all","sort":"DESC","perPage":30,"cursor":"djE6ks8AAAAEuPRpAwA","startCursor":null,"endCursor":null}},"title":"Activity · mountaineerbr/shellChatGPT"}