I recently had to do a complete reinstall of jupyter on my machine. Since then, when I use nbextensions to convert my notebooks to an html file, the table of contents links to the original notebook rather than the new html file. Before, the basic command jupyter nbconvert --to html mynotebook.ipynb produced a functional table of contents.
Unfortunately, I have no idea why toc2 is no longer properly converting to HTML. I will appreciate any insights that anyone can give me.
It turns out that it was a bug introduced to toc2 by a recent update. toc2 has since been updated to fix this issue.
Related
I'm using VSCode. After the last update of the IDE, I see this error (Even if I didn't open any project to work on).
I googled it, and I tried several actions according to this link:
https://github.com/microsoft/TypeScript/issues/43838
but none of them seems to work.
Anyone faced the issue? (I'm not using TypeScript in any of my projects and I don't have any tsconfig.ts file). Is it related to one of my VSCode extensions?
It is a general update problem about visual-studio-code with the last update. Issue can be followed from here: https://github.com/microsoft/vscode/issues/168420
Installing JavaScript and TypeScript Nightly extension is helpful for me.
Configure .eslintrc.json in your folder again and reload the vs code
example:
.eslintrc.json file for salesforce
Or
you can refer below link to config your eslint file
https://eslint.org/docs/latest/user-guide/configuring/
I have been using Code-OSS in manjaro linux for some months now and this is the first time I am encountering this. (After I reinstalled manjaro) When I search for e.g, Java Extension Pack or Intellisense I don't get any result in the extension marketplace. I don't know if it's the problem of my setup or everyone is experiencing this. If you know of a solution please tell me.
Thanks in advance
This can be fixed by adding following to product.json:
"extensionsGallery": {
"serviceUrl": "https://marketplace.visualstudio.com/_apis/public/gallery",
"cacheUrl": "https://vscode.blob.core.windows.net/gallery/index",
"itemUrl": "https://marketplace.visualstudio.com/items"
}
This can also be fixed by copying product.json from an official build (in the .zip archvie, product.json is under /usr/share/code/resources/app/product.json which contains above lines). Depending on your distribution, exact location may vary.
More information: https://github.com/Microsoft/vscode/issues/1557
As an alternative to manually editing product.json or building whole visual-studio-code-bin from AUR you can use different AUR package - code-marketplace - that patches product.json from Code OSS package (code)
Thank you #Clay for responding. I saw that post prior to me posting this question. Those bits of json code were already in my product.json file but still I was not able to get the extensions. Then I built the VISUAL STUDIO CODE (visual-studio-code-bin) from AUR and now its working normally as expected. I don't know what's the reason behind this,but that's how I fixed it.
Also if someone is reading this and wanna know about the different variants of VSCode then here is the archWiki page for VSCode.
For anyone using the lscr.io/linuxserver/code-server docker container like me, the .json is located at /app/code-server/lib/vscode/product.json.
I've encountered this using OSS Code on Manjaro as well, trying to install the Azure Resource Manager Tools extension. Not sure if this was an option when this question was posted, but you can download the extension (.vsix) file if possible, then bring up 'quick open' in OSS Code using Ctrl+Shift+P, search for "VSIX" and choose "Extensions: install from VSIX" to locate and install the downloaded .vsix file.
After this, the extension was installed and usable in OSS Code, and is also configurable in the extension manager panel.
I need all codecells in a jupyter lab document to run when opening it in binder. I've tried adding jupyter trust <filename> to the postbuild file but it does'nt work. Any tips are welcome.
When you say that it doesn’t work, does it at least show as trusted when you open the notebook? Because that approach is/was recommended way to set to trusted. Until the extension enabling example notebook was streamlined to only include an example that works in both the classic an JupyterLab interface, that was the way documented here. Here discusses how to check if trusted.
However, just setting trusted does not trigger the code to run when the notebook is launched.
Aside of coding extension that will run code like python-markdown, there is no simple way to trigger that in the Binderhub system at this time. When this question came up the other day in the binder gitter channel, I suggested a person could most likely make an empty notebook appear to have just been run by using a start config file to trigger execution of the notebook via nbconvert or papermill.
Alternatively, using Appmode as discussed here will execute all cells when run in app mode; however, the interface won’t be a notebook. Similarly, installing Voila and linking to the Voila render will run code in a notebook at launch. It will look like a web app interface and not a notebook though.
Why do I have the same suggestions while typing expression?
Example:
I had exactly the same problem. After a week or so, it get really annoying.
basically, as the comments hint to, there are probably multiple linting or intelisense tools. In my case (for python) i had the pylance extension added.
When i disabled this, the problem went away, but features were missing. So i added it back...
For some reason (i dont know why), this fixed the problem !!!
I can only hypothesise that in some way the extension was corrupted. Nevertheless, it worked.
EDIT:
I can also confirm that unchecking this setting appears to work:
Jupyter: Pylance Handles Notebooks
My current system is Windows 11, with python 3.10.
Final edit (8 Dec 2022):
This is resolved here:
Please could you install VS Code 1.74 and the latest Jupyter, PyLance and Python extension and confirm this still exists.
Visual Studio Code provides an API so third-party extensions and built-in modules can contribute suggestions for auto-completion pop-ups. The system is currently designed so suggestions are merely appended—there's no duplicate detection or removal (perhaps because extensions can also take care of sorting suggestions and such algorithm would get on the way). That means that if you have more than one extension or module for a given language you can easily get duplicate entries.
Having several extensions for PHP is not necessarily a bad idea since they can address different needs (for instance, PHP DocBlocker just creates annotations, it doesn't provide auto-completion suggestions) but you have at least two extensions (PHP Intelephense and PHP Intellisense) that do exactly the same things. That's likely to hurt performance (all your workspace files will be scanned several times) and just adds noise.
I suggest you read the extension descriptions carefully to learn what they do exactly and then figure out which ones you need. Remember that extensions can be enabled/disabled in a per-workspace basis.
The following is just my own totally subjective opinion. Among the PHP extensions that provide code intelligence only two of them seem mature enough:
PHP Intelephense
PHP Intellisense
I've tried both. PHP Intelephense works best for me than PHP Intellisense so that's the one I've kept. I've also disabled php.suggest.basic following the installation instructions because basic suggestions didn't add any value to me (they were blind string matching):
Turn off the php.suggest.basic setting for best results.
... as well as taming builtin Emmet support, which was providing really dumb suggestions:
"emmet.showExpandedAbbreviation": "inMarkupAndStylesheetFilesOnly"
YMMV.
TLDR; Installing pre-release version of Jupyter solves (v2022.11...)
Ok, so after some more extensive experimentation I think I found what's causing this in my case. After looking at the processes I noticed that there were two Pylance processes running, and consistently this would only be a problem if I was working in a session with a jupyter notebook open or one that had been opened.
saun89 17740 37.3 0.3 1008004 199492 ? Sl 20:58 0:22 /home/saun89/.vscode-server-insiders/bin/fef85ea792f6627c83024d1df726ca729d8c9cb3/node /home/saun89/.vscode-server-insiders/extensions/ms-python.vscode-pylance-2022.11.32/dist/server.bundle.js --cancellationReceive=file:9178e897a2b78b36bfd167f79b36c3bdad2931d71b --node-ipc --clientProcessId=17651
saun89 18743 257 0.7 1304584 382288 ? Sl 20:59 0:20 /home/saun89/.vscode-server-insiders/bin/fef85ea792f6627c83024d1df726ca729d8c9cb3/node /home/saun89/.vscode-server-insiders/extensions/ms-python.vscode-pylance-2022.11.32/dist/server.bundle.js --cancellationReceive=file:8744a321767eed92821fd737be4dc7dcfb728284e5 --node-ipc --clientProcessId=17651
Pylance basically spins up a service for the workspace, and then spins up a separate service for the notebook.
Output from "Python Language Server" logs:
Disabling Jupyter removes the duplication, and after installing an earlier version of the extension (v2022.4) this appears to have fully resolved the issue. I'm going to go ahead and log the extension bug once I have something reproducible.
As of 11/30/22, Jupyter Extension Pre-Release version v2022.11.1003281132 is the latest version fixes this issue. Click the gear icon next to the extension and you should see "install another version..." Then you can select version v2022.11.1003281132.
So, I've been using Sublime for a while as a simple text editor, but I'm venturing into the land of plugins for work and I've run into an issue getting pylint to work. Specifically I have it installed and have the Sublime package manager working, but I'm not sure how to include paths in my sublime settings.
I haven't found very useful documentation on this point, but if you are willing to point me towards it that would be a perfectly acceptable answer. My basic issue is that currently every time I save a file, the following error message shows up:
"Please define the full path to 'lint.py' in the settings"
I understand this error message which is great, I just have no idea where the sublime settings are or what the standard format for defining a path is. Any help would be appreciated.
There are quite complete directions at the SublimeLinter GitHub page that should describe how to set everything up. Briefly, go to Preferences->Package Settings->SublimeLinter->Settings - Default to see where the different settings are defined. Then, open Preferences->Package Settings->SublimeLinter->Settings - User to change anything, as the main settings will be overwritten when the package is updated. Remember that settings files are in JSON format, so you'll need opening and closing curly brackets to enclose settings, and a comma between them:
{
"sublimelinter_delay": 30,
"sublimelinter_mark_style": "none"
}
I haven't gotten the error you have, are you sure you're using the regular SublimeLinter and not the new beta version? I just installed it fresh on my work machine (I've been using it for a while at home) and after restarting ST2 it's working like a charm with Python.
I had the same problem. I found it to be coming from the package "Pylinter". I removed Pylinter and added SublimeLinter, it covers more than python and is well used. I'm enjoying it quite a bit.
From your command line just run:
sudo easy_install pylint
sudo easy_install pep8
After that restart your sublime.
I encounter the same situation today. In my case, that is because I have not installed the 'pylint' in my system yet. It works fine after I installed the 'pylint' through pip.
Just for other people reference.