I'd like to be able to provide a regular expression for text to ignore when creating diff's with sublimerge. Is that possible?
From Borys, the author of sublimerge:
Please take a look at http://www.sublimerge.com/sm3/docs/configuration.html#default-settings-diff-algorithm - there is a „intraline_unimportant_regexps” setting that should do what you want, however this is not a live setting so it requires the diff to be re-processed.
In sublimerge3 (2?) there is a "Recompare buffers" option in the "sublimerge" section of the right click context menu (when you're viewing a diff). Also F5 will re-process the diff.
Note that at present this only changes the highlight on "unimportant" differences, they still appear when you 'go to next diff'.
Related
See the "..." before "layout-1c"? Firefox seems to have some rules defined for when it should collapse a lengthy element attribute. I want to disable this. Just show me everything, always. How?
EDIT double-clicking the attribute is not good enough. It loses wrapping and makes you manually scroll through a ton. Firebug does everything right from the get-go and I'd like the same behavior for the built-in inspector.
In the official mozilla docs it explains how we can tweak whether to truncate or not, and also how many characters will be shown before being truncated:
Truncate DOM attributes
By default, the Inspector truncates DOM attributes that are more than 120 characters long. Uncheck this box to prevent this behavior. This setting works by toggling the about:config preference "devtools.markup.collapseAttributes". To change the threshold at which attributes are truncated, you can edit the about:config preference "devtools.markup.collapseAttributeLength".
There is a convenient checkbox in developer tools.
To find it click the three dots in the upper-right corner and go to Settings:
Truncating long attribute strings is, so far, a feature of the Firefox inspector, as in some cases, attributes may be really long and make it hard to use the tool. Think of base-64 image data-URLs for example.
I understand that this feature might not always be wanted, but for this to change, the inspector's code needs to be changed, there isn't a setting you can use. 2 options:
Increase the limit after which strings are truncated to make sure only really long attributes are,
Or add a setting (off by default to preserve today's behavior) to turn this off entirely.
But to answer the original question, no you can't disable this (other than by changing the code, or writing an addon that would monkey-patch this).
You can double-click the attribute value to show the collapsed content. Note that this is a non-persistent solution.
The most optimal workaround speed-wise seems to be right-click "Edit as HTML".
Fortunately we have a bug filed now https://bugzilla.mozilla.org/show_bug.cgi?id=1198073
In VS2013, I find it irritating that it jumps wildly through the current source file to places matching my incomplete entry, while I am typing - I would prefer if it did nothing at all until I completed my entry and hit Enter (and if it finds nothing, the scroll position in the file will not be changed).
I couldn't find an option under "Tools/Options..." that looks like it helps there.
Is it possible to do this?
No, there is no user setting for that, sadly.
Instead, however, consider using SHIFT CTRL+F (find in files). Not the obvious choice, but I've started to prefer it as it gets rid of the leaping about and it can anyway be useful to see the list of occurrences before losing scroll position (whether or not it finds anything).
When I extract a method in CodeRush, it has some default formatting. I'd like to change it. Specifically, when I choose the location of the extracted method, CodeRush smashes it in, directly above the method I put the location marker at.
So:
When I choose to drop the extracted method, like this:
It ends up looking like this:
What I want is for it to have some control over how the method looks, at the moment I want to add some blank lines between it and the next method. How can I do that?
Open the DevExpress -> Options dialog;
Go to the Editor | Code Formatting | Blank Lines option page;
Toggle the 'After multi-line members' option.
I requested this here: http://www.devexpress.com/Support/Center/p/S130722.aspx
It got implemented here: http://www.devexpress.com/Support/Center/p/S19229.aspx
There's now a massive formatting subsystem I spent 3 hours fighting that tries to outdo ReSharper's massive formatting subsystem which both need to be able to survive VS Autoformat (CtrlK D) and don't always.
Now you know it's there, I'd appreciate a quick summary from you as to whether you were able to make it do what you wanted!
Can anyone tell me if there's a quick way to format your code in Text Mate, similar to pressing ctrl K+D in Visual studio?
Thanks!
Edit by Damien_The_Unbeliever:
For those not familiar with Ctrl K+D, it doesn't just indent code - it reformats it using the generally established formatting conventions in the editor - it may replace spaces with tabs or vice-versa for the indentation, ensure code is consistently indented, move braces to separate lines, etc.
TextMate reindenting and reformatting varies a little depending on the language you're using.
You can generally use the Text menu, that depending wether you have an active selection or not it will show you different commands under it. For example, if you have selected a section of code, there will be a Indent Selection menu item. If you have no active selection, it will be Indent Line.
To have this working properly, be sure to select the current language, if it isn't assigned yet (like on unsaved documents). If you're working with HTML, it will simply indent the lines depending on what's above it. It will keep line breaks intact.
If you need something to break out tags on new lines and properly format the document, you should use the Tidy command that is found in the Bundles menu, under HTML (or simply by using the shortcut CTRL+SHIFT+H. If you have a selection active at the moment that you use it, it will simply reindent that section. If instead you have no selection, it will properly reformat the whole document, including checking for tag validity and other errors.
The Bundles for other language have similar commands, like XML (still Tidy) and Javascript (that has a Reformat Document command).
As an ending note, I simply suggest to look into the Bundles menu; there are many little gems in it. ;)
Did you look in the menu bar? Under Text you have a couple of Reformat… entries that may fit your needs.
Beside these native features, some bundles — like the JavaScript one — have custom Reformat… commands : click on the little cog button at the bottom and explore your current language's bundle's content.
When I compare 2 files with WinMerge on Windows platform, do you know how could I ignore specific words ?
I mean I wanna say to WinMerge : "Don't care about that list of words : house, garden, etc."
An so when comparing those 2 lines :
the house is at london
&
the bus is at london
then no difference will be displayed since I don't care about the word house...
Any idea ?
In winmerge 2.16 (at least) (2020-05), you can use menu Plugins > Plugin Settings...
Click on PrediffLineFilter.sct, then click on the "Plugin Settings..." button, check in "Use RegExp" check box in the corresponding column. In the "Find what" column, specify a regexp and leave the "Replace with" cell empty. Close this plugin configuration window. Then, in the Plugins > Prediffer menu, select PrediffLineFilter.sct.
You can then turn off this filter plugin using menu Plugins > Prediffer > No Prediffer (normal).
I know its been a while since this was asked, but since this shows up as a top match when searching for this topic, I thought I'd add to it...
As far as I can tell, WinMerge STILL doesn't have this functionality directly. You have 2 options:
1) Write your own plugin for WinMerge. It does seem to be possible to do what you want that way if you wish to invest the time.
2) Get Meld. It includes the ability to use regex to ignore the parts of a line you don't want compared. For example, if you just put "word" in the text filter preferences it will ignore "word" and compare all other parts of the line. I use this to compare dmesg logs in Linux, ignoring the timer field at the start with "^[.*?]" (ignore everything between the initial brackets, non-greedy).
I am looking at the moment for something similar, so perhaps this may be helpful for you:
WinMerge does not allow to filter parts of a line. If a line filter matches, the whole line is ignored. See http://manual.winmerge.org/Filters.html for an explanation of line filters.
So if there is a difference other than the word bus and house in it, the line will be ignored as well.