An Option to keep the search box open and closed only when we use the close button would be very useful when we are searching in many tabs.

Regards

This matches Chrome's native Find feature, at least on Linux.

The normal search of Chrome is able to search in opened PDFs.

It would be nice of if the regex search would also be able to do that.

Some sites use iframes to show main content, but searching inside them not possible.
It will be very useful if your extension will support searching text in iframes.
Will you plan to add this functionality?

Roger, would it be possible for the search form to open with previously-entered text selected in the input field? That way, if I had a new regex to enter, I could type it without first having to clear the field.

Example: I have v1.0.2 installed and mapped to Ctrl+Shift+F. When I hit the hotkey and enter a regex, everything works fine. I then hit Esc to close the search box, and hit Ctrl+Shift+F again. The text I entered is still there (which is fine), but the text is not selected. In order to start a new search, I have to clear the box first.

There is an easy workaround, of course - Shift+Home, Del to clear the box :) . However, I also use Chrome's built-in find (Ctrl+F) for plain-text searching. The behaviour I am requesting is the same as that of the Chrome find. The consistency between the two would make it much easier for me to use Regex Search without regular "Oops --- forgot to clear the box!" moments.

Thanks for considering this request!

Chris White

Aka white text on black background.

Setting color: black; on highlighted text should fix this.

Having just installed this Chrome extension, I was surprised that there were no mention of the supported RegEx syntax. Neither on the Chrome Webstore, here on the ReadMe, in the extension itself (Right-click>Help), nor is it mentioned in an issue here on GitHub.

It would be REALLY useful with a compact "cheat sheet" for those who do not recall which character class shortcut to Word Break, how to express a Reluctant quantifier, or specify negative lookbehind.

Also, from within the extension context menu, it would be useful with a link to a complete syntax reference for the RegEx dialect.

Provide a mechanism to allow filter the tags the code will apply regex.

For example, consider the site containing the GNU BASH manual https://www.gnu.org/software/bash/manual/bash.html

Suppose you want find the doc of "declare" builtin. Then, you search for "^declare$"., in order to find the section where contents about "declare" are. Currently, the RegExp extension searches 27 matches.

This is due fact that inside the page, there are many "declare" words inside tags like "code", "var", etc. This is nice, because our search finds reals mentions to the "declare builtin" or some code about it, not only a "declare" simple word...

But, we want a way to search a "declare" that is not middle of a text. For example, if a can specify the "^declare$" apply only to text nodes inside "dt" tag, then, i could find the desired declare. Or better, if i can exclude the any text node inside a "var", or "code" tag, this is useful also.

It is not working on Chrome now. The extension address.

How can I use a shortcut/hotkey to copy the current matches to the clipboard?

Hi. I love it! I've set it to Ctrl-F. :)

I do both case sensitive and insensitive searches. Having to pull up the Options page to switch between the two modes is a bit of a nuisance. I wonder if this option could be moved to the search box instead?

The only thing I miss, compared to Chrome's search, is marking the finds on the scroll bar. Any chance of this? Even if it meant adding a bar next to the scroll bar, I think it would be handy.

Thanks!

Any AWS console page breaks this plugin:
eg:
https://eu-central-1.console.aws.amazon.com/ec2/v2/

Incorrect highlighting or reported number of occurrences found. Hard to characterize as it appears that word delimiters on the AWS console break this plugin.
Not including any screenshots as I'd have to redact a lot, but it's clearly not working properly when used on any of the AWS console ec2 pages.

Nice extension! Would you please update the manifest (and source?) so it will also work on file:// URLs? My use case involves a lot of locally-stored HTML.

I have not done any Chrome extension development myself, but found this doc page and this SO question* that suggest adding "file://*/*" to the matches in manifest.json should do it.

Thank you!

*Stack Overflow question above by Anthony

Hello,
Maybe related to #60 since SVG is practically generated content, but maybe different.

I have a locally stored tool that practically does the following, all in Javascript:

• HTML placeholder page having 3 frames
• Each frame uses JavaScript to load an XML file and applying an XSL template.
• Result is two navigation panes, and a content pane.

No matter what I search for, it may even be ".", I get 0 hits.
Yes, allow access files from URLs is enabled. Otherwise, there was a big red warning that this page couldn't be opened.

Question: is the tool supposed to find things:
A) on the page, (contains the contents of the XML file loaded.
B) or in the page source? (contains only javascript, no content)

When you just want to search for something new, hit the shortcut and start typing, it feels "wrong" to then realize that what you typed was simply appended to the last search and now you have to go and manually remove that part.
A boolean option to enable selecting the input's contents on activation would be perfect:

• Simple to implement
• People who like the other way better can continue to not use it
• People who want a feel more akin to the native Ctrl+F behavior are happy 😊

Today for the first time https://bugs.chromium.org/p/chromium/issues/detail?id=490915 negatively affected me (version below).

Steps: Run a regex search from the popup on any Internet page for text on that page

Expected: Search result count listed in the popup

Observed: Result count stays at 0 of 0

Request: Would you be willing to update the search function to test for the case described below, warn, and still perform a search using the default settings? Thanks for considering this request!

Analysis

Here, the extension (reasonably) assumes Chrome will behave as documented :) . However, Chrome instead returns undefined.

MCVE (in the developer console, with "Chrome Regex Search" selected, after the above error occurs):

chrome.storage.local.get(null,(r)=>{console.log(chrome.runtime.lastError); console.log(r);})
undefined
{}
    message: "IO error: .../LOCK: Access denied. (ChromeMethodBFE: 15::LockFile::5)"
    get message: ƒ ()
    __proto__: Object
undefined

Browser

Chromium 76.0.3809.87 (Official Build) snap (64-bit)
Revision 111fe1e15d5ced26080a7dc239bcfe70f6c49aad-refs/branch-heads/3809@{#967}
OS Linux (Ubuntu 18.04.2 LTS)
JavaScript V8 7.6.303.27
Flash (Disabled)
User Agent Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) snap Chromium/76.0.3809.87 Chrome/76.0.3809.87 Safari/537.36
Command Line /snap/chromium/813/usr/lib/chromium-browser/chrome --no-default-browser-check --no-first-run --password-store=basic --profile-directory=Default --flag-switches-begin --disable-smooth-scrolling --disable-features=HardwareMediaKeyHandling --flag-switches-end
Executable Path /snap/chromium/813/usr/lib/chromium-browser/chrome
Profile Path /home/cwhite/snap/chromium/813/.config/chromium/Default

When I use a keyboard shortcut to open Chrome Regex Search and immediately start typing a query string, the first few characters in my query string are lost. Sometimes, the first character or two end up in a text field on the web page, even though I type them after the keyboard shortcut. I don't mind that the dialogue takes a little while to appear (a few hundred milliseconds), but I'd map this extension to Ctrl+F if none of the characters I typed got lost.

Chrome 65 under Linux Mint 18.1 64b (Cinnamon 3.2.7) on an admittedly old laptop (dual core i7 @ 2.90GHz).

Hello.
It seems it doesn't work in this simple web:
https://cran.r-project.org/web/packages/
It just doesn't find anything.

How do I select with multiline or is it not supported? thanks

I am trying to search for for loops in some python code examples for a module.

I can see one on the screen so I know where it should be highlighted.
in the search bar I use:
for\s\w+\sin
to search for "for" followed by a whitespace character, then any variable name, followed by another whitespace character and finally the word "in"

This formula works for text on the page that is not formatted as code, but not if it is in a code block.

The extension CAN find single words in code blocks, if I search for just "for" it manages to find them in code blocks (as well as outside of them)

in fact white space in code formatting seems to completely block this extension (even when using .* to select ALL characters following the searched string):

Here I have searched for all examples of the characters "de" and then everything that follows them. as you can see, the white space in the code blocks ends this search.

Interestingly there is no issue with chrome's standard Ctrl+F function, that can find full sentences inside of code blocks.

It would be nice if we can use regex search in pages with forms. In other words...find matches within the content of text input or text area fields (maybe even select options).

I'm sure it's common for people to need to find fields containing one or more values in order to change them. In my case, I am updating a product with either of two color options.

I'll try to clone this project and give it a try at this if I can find some time, but wanted to open this feature request first.

System: Host: linux-fcke Kernel: 4.12.8-1-default x86_64 bits: 64 gcc: 7.1.1
Desktop: KDE Plasma 5.10.4 (Qt 5.9.1) dm: sddm,sddm Distro: openSUSE Tumbleweed 20170822

Information for package google-chrome-beta:

Repository : google
Name : google-chrome-beta
Version : 61.0.3163.59-1
Arch : x86_64
Vendor : Google Inc.
Installed Size : 343.7 MiB
Installed : Yes
Status : up-to-date
Source package : google-chrome-beta-61.0.3163.59-1.src
Summary : Google Chrome (beta)
Description :
The web browser from Google

Google Chrome is a browser that combines a minimal design with sophisticated technology to make the web faster, safer, and easier.

using extension https://chrome.google.com/webstore/detail/chrome-regex-search/bpelaihoicobbkgmhcbikncnpacdbknn/support?utm_source=chrome-app-launcher-info-dialog

tested on this site https://cgit.kde.org/konversation.git/commit/?id=72de2027f3d3e7a0c48aef6acd666e04fc0b19b1

Regular Expression patterns should be able to use ^ and $ characters to denote the start and end of the line, respectively. However, these characters cause the pattern to not match at all.

I'm trying to find text on a page that uses the HTML ^{tag for superscript and the add-on is not able to search for text which spans the sup tag.}

For example this page (it's quite large): http://oracc.museum.upenn.edu/cams/gkab/cbd/akk/onebigfile.html

If you Ctrl-F "dubdub" in the normal Chrome search you find four text matches, however when you do the same search with the add-on you only get two. The two missing from the add-on are strings where half of the search term appears in superscript.

initial ^ not working in txt files, like http://data.iana.org/TLD/tlds-alpha-by-domain.txt

Is it possible to add Firefox support ?

My main reason for wanting this is to search for things in the AWS console. Unfortunately, CloudWatch Logs (and many others) use iframes. Is there a security limitation that would prevent you from accessing frames?

This is a really useful extension. User will need a function to copy those selected text. Is there any way to do this?

Please add an option to search no matter of accents.
Something like the "case insensitive" option but ignoring accents instead of capitalization.

For example I would like to find both "último", "ultimó" and "ultímo" by just searching for "ultimo".

Although the standard 'find next' and 'find previous' shortcuts in Chrome for Linux and Windows are Ctrl+g and Ctrl+Shift+g, the old standards of F3 and Shift+F3 also work. They ought to work in Chrome Regex Search too.

If you try to search for test here then only the regular variant will be found. The normal search of Chrome handles this case as one would expect.

regular: test
bold serif: 𝐭𝐞𝐬𝐭
bold sans: 𝘁𝗲𝘀𝘁
italics serif: 𝑡𝑒𝑠𝑡
italics sans: 𝘵𝘦𝘴𝘵
bold italic serif: 𝒕𝒆𝒔𝒕
bold italic sans: 𝙩𝙚𝙨𝙩

The pattern "^some text$" dont works and matches "some text" in middle of line.

It looks like if I copy Japanese text and paste it directly into the search, it works. (test: copy this into the search: ひらがな)

However, if I type the Japanese directly into the regex search box, it comes up with zero results. (I guess to test this, you need to be able to type in Japanese)

cmd+F only launches the embdedded search, I have to click on the extension everytime, please help.

When the search box is open.
E.g. can't switch tabs (Ctrl + Tab), close tabs (Ctrl + W)

The search is unable to and digest escape sequence.

One example will be to search on the Java API page. When I want to find java.lang.Thread, I search for '.Thread$', but it does not work. The period character isn't escape in the search bar.

Often, when we search something in Chrome, we mean to select it for copy paste or to click the link. In Chrome, when we search for something and then hit escape, we have the found entry we were on selected and we can just hit Ctrl + C. Unfortunately, this is not the case here.

• The text is just highlighted and not selected. As such, I would like to propose that we modify the searcher to make it select the entry we're on instead of highlighting it in a different color and to remove the highlighting when we exit the searcher via Esc.

• Also, in Chrome, when we hit Ctrl + Enter while we've selected part of a text via Ctrl + F, it has the same effect as clicking that selection. This would also be a very useful feature to have.

Reproduction:

1. Search for something with multiple results.
2. Press Shift.
3. Press Enter
4. Release Shift.
5. Release Enter.
   The order is important. If you mix up 4 and 5, it will work correctly.

When you press Enter, you'll go to the previous result. Ok. But when you release Shift, you'll go forward again.

Here is a screenshot of the issue:

I am using Vivaldi browser which is based on Chromium. Regex searches appear to work as they do in chrome, but setting a keyboard shortcut does not. I have it set to ctrl+shift+f and it does not bring up the search dialog. Also, when clicking on the icon, escape does not exit out of it.

This is an awesome add on!

Bug #2 added support for file://, but for me all of my files start with file:/// (with three /, not two) Could you add support fir that? I am not sure if file:// is even a thing as I can not use it - google chrome automaticly adds the third /. I would imagine this would be a quick fix, probally one line. Do you mind adding support for file:///?

https://projects.lukehaas.me/regexhub/
Useful regex can get really long and annoying to write, thus it would be cool if the extension provided small distinguishable icons/buttons for example that fill the regex field (search field) with some regex template code.

I think instead of individual icons, however, it would be easier to just show a list of text, since some regex is pretty specific and hard to translate into an icon.

On the following US senate vote webpage, I was trying to match the democrats who voted Yes with "D.*Yea" but your tool does not find any, maybe because of the bold tags around the Yea.
https://www.senate.gov/legislative/LIS/roll_call_lists/roll_call_vote_cfm.cfm?congress=115&session=1&vote=00195

I found the votes grouped by position further down the page, so I could answer my question, but in general, a more robust matching of text through HTML formatting would be nice.

On Chrome Version 87.0.4280.88 (Official Build) (64-bit)

regex extension is working great with nfl.com,
but when I try to use it with a local html report file I get this error

Content script was not loaded. Are you currently in the Chrome Web Store or in a chrome:// page? If you are, content scripts won't work here. If not, please wait for the page to finish loading or refresh the page.

if I load the same file to my webserver it works great as well
https://www.securedata.cadenhance.com/spreadsheet_html_report/html_spreadsheet_report_01.html

Can you download the file from the site above and try looking at a local copy?

Ideally, add a toggleable button like [Aa] on the search bar to allow case insensitive search. Also add option in settings for case-insensitive search to be default.

search 'john' in https://mermaid-js.github.io/mermaid/#/./sequenceDiagram
words in code blocks are highlighted, but words in sequence diagram are not highlighted

I wanted to be able to find the number 70 in the image below:

The problem is:
Repeated pressing in the search text box or reselecting the same history record is not applied (doing nothing). It's necessary on interactive pages, e.g. on log pages, since after refreshing the page, the highlighting of the search results will disappear.

Here: https://www.listediparole.it/m/a/4/paroleconario.htm

You can try this regex:
.{1}ario
no word from the results will be highlighted.

It is probably because the formatting makes the searcher believe that each word result is actually two different words (separated by <b> tag). So that they are individually searchable.

Default Chrome search does not do this.