_mk_
-
Posts
163 -
Joined
-
Last visited
-
Days Won
5
Content Type
Blogs
Gallery
Downloads
Events
Profiles
Forums
Articles
Media Demo
Everything posted by _mk_
-
No problem with @ and emoji in my machine. Do you use the latest version 2.3 of the workflow?
-
Assigning different keywords in the workflow has great potential to break certain functionality. Selecting a project/context/perspective from their searches (.offp, .offc, .offv) will invoke the corresponding .oft* keyword. The same keywords are used for the back functionality of the task details display. So changing the keywords is not a good idea unless you also change all other related scripts. Nested contexts are currently not supported. The workflow will only display tasks that are directly in the searched context. So, parent contexts will not show the tasks of their child contexts. This will be fixed with the next release as I already had to implement the recursion for the folder search. Don't you get any tasks for @ and Emoji contexts don't they even show up in the context search (.offc)?
-
Could you please try to cd in the workflow directory first before executing the command? cd /Users/emory/Dropbox/Application\ Support/Alfred/Alfred.alfredpreferences/workflows/user.workflow.E1A1A3ED-90BF-4B96-BD56-16B26C7193D3 ./getTasksInInbox.sh Do you still get errors?
-
What are the currently available tasks in this context? Does one of them maybe contain any chars other that letters and digits?
-
The fallback web searches appear if no results are returned from the workflow. But when your context says that there are available tasks, they should show up. Could you please try to start the context script in Terminal: ./getTasksByContext.sh "Phone" Do you get any error messages? I'll add some logging with the next release for better support in the future.
-
When using .oftc (OmniFocus Tasks in Context) you have to enter the complete context name. Use .offc (OmniFocus find Context) instead. When you start typing the first letters, you'll get a list of all matching contexts. Select the right one from the list and it will call .oftc with the correct context name.
-
Thanks. The folder search is already on my todo list for the next release. Opening the project in OF is currently not possible but should be simple to add. Keep an eye on the next release.
-
I just checked and the links are ok. Could you please right click and use "save as" to download them.
-
It will switch to the fallback websearch if no results have been found. Which keywords are working for you and which don't?
-
I understand that it would be too much of an effort for Andrew to support multiple languages in Alfred itself. But that was not my suggestion. I could imagine an e.g. Script Filter edit screen that offers the possibility to define placeholder texts for different languages. The same applies for notifications. I also saw Vero's suggestion to use different keywords for localized searched/script filters. But that would bloat the workflows as one would have to add script filters for every supported language. The other solution would be to offer different workflows for different locales. But that would also unnecessarily create a lot of effort for the workflow developers. I am not quite sure if Alfred is only targeted to power users that would not like to have workflows in their native language. At least, I would like this.
-
It is easy to get the current system language of the mac and create feedback xml in multiple language. Unfortunately, it is not possible to define "Placeholder title", "Placeholder subtext" etc in different languages. Would it be possible to add multi-language support for the strings that can only be modified in the workflow editor?
-
This is going to be a busy evening... I almost gave up on fixing this issue but just a few minutes after releasing the last version, I found a solution to the problem. Sorry for the inconvenience. Version 2.3 fix no results when searching for german umlauts (and maybe other accented characters Download
-
I am having a similar issue with german umlauts. I was able so solve the issue with the normalise command line tool posted by Andrew in another topic. You could add Andrews tool to your workflow and try to do the following: QUERY=$(./normalise "{query}") DATE=$(echo "${QUERY}" | sed -e's/HAEC/CEST/g' -e's/HNEC/CET/g'); export LC_TIME="fr_FR.UTF-8"; date -jf "%d %B %Y %T %Z" "$DATE" "+%Y%m%d%H%M%S"
-
I also had problems with the expansion of ~ in my workflows. Instead of ~ use $HOME.
-
There was a bug in version 2.2 caused by the updated bash workflowHandler. Please update to the latest version 2.21
-
I made an important update to the workflowHandler: When generating the feedback XML special chars are properly escaped. Trying to save preferences with the & char caused weired behavior on the shell Everyone using the workflowHandler should update to the latest version.
-
Thanks to CrashPlan I was able to recover the essential parts of my profile very quickly. So I can release version 2.2 of the workflow with the new features that did not make it into 2.1. Version 2.2 new task actionsadd/append note (thanks to uscmeche) change project change context hotkey support​SHIFT+CTRL+I - Show tasks in Inbox SHIFT+CTRL+D - Show tasks done today SHIFT+CTRL+P - Search for project SHIFT+CTRL+C - Search for context SHIFT+CTRL+V - Search for perspective Bugfixes Download
-
So, Alfred now also learns Script Filter preferences?
-
The clip.png was in there for that purpose. I'll include your addition in the next release. I am still trying to display the note in the task details but that's giving me a hard time as the note is in rich text format.
-
Version 2.1 is available: search for perspectives in context, project and perspective search all contexts/projects/perspectives are shown if no search string is entered the flagged status of a task is visible in all task result views the database is queries instead of the project metadata to find projects to improve performance task details link to project and context view task's start and due date can be deferred from task details view This morning I dropped my MacBook on the floor and crashed my second data HDD. Fortunately, the system still runs fine from the SSD and only my profile got lost. So the next update will take a few more days until I recovered my data from the CrashPlan backup. I hope you enjoy the update in the meantime.
-
I simply do the following to generate random uids: T=$(date +"%s") echo "<item uid='${T}n' ...>! n is just incremented for each result. This should not create any noticable delay.
-
Why don't you split the playlists (maybe one file per playlist) into multiple json files and loop over them?
-
You can get the non-python version here.
-
I'll upload my modified version of the workflow this evening.
