How can I use Automator to extract substring of text based on pattern?

If your pinyin substring never has spaces in it, then you can use AppleScript to extract it like this:

This is the script (with the sample input). Copy/paste into AppleScript Editor and click the green triangle 'Run' button:

set input to "我 wǒ 代 ① （指一人） （用作主语） I （用作宾语） me （表所属关系） my 告诉我 tell me 我为人人，人人为我 one for all and all for one 我爸/妈 my father/mother 我的祖国 my homeland 我现在没空。 I am busy at the moment. 我认为我行！ I think I can manage it. ② （指两人或以上） （用作主语） we （用作宾语） us （表所属关系） our 我厂/国/校/军 our factory/country/school/army 敌军被我全歼。 The enemy was annihilated by us. → 我方, 敌我矛盾 ③ （表泛指） [used together with 你 in parallel structures] anyone 大家你一言，我一语，献计献策。 They had a brainstorming session with anyone and everyone joining in. 市场里你来我往非常热闹。 The market is bustling with people coming and going. → 尔虞我诈, 你死我活 ④ （指自我） self → 忘我, 自我"

set pyCharSet to {"a", "ā", "á", "ǎ", "à", "b", "c", "d", "e", "ē", "é", "ě", "è", "f", "g", "h", "i", "ī", "í", "ǐ", "ì", "j", "k", "l", "m", "n", "o", "ō", "ó", "ǒ", "ò", "p", "q", "r", "s", "t", "u", "ū", "ú", "ǔ", "ù", "ǖ", "ǚ", "ǜ", "w", "x", "y", "z"}

set {oTID, AppleScript'stext item delimiters} to {AppleScript'stext item delimiters, "①"}

set cc to (input as string)'s text items's item 1's characters 2 thru -1

set py to ""

repeat with c in cc

if c is in pyCharSet then set py to py & c

end repeat

set AppleScript'stext item delimiters to oTID

return py

The above view is AppleScript Editor, not Automator. What do you plan to do with the results in Automator?

SG

Here is an abbreviated way to capture wŏ from your input stream. Note that AppleScript wants to escape certain characters.

set inbound to "我 wǒ 代 ① （指一人） （用作主语） I （用作宾语） me （表所属关系） my 告诉我 tell me 我为人人，人人为我 one for all and all for one 我爸/妈 my father/mother 我的祖国 my homeland 我现在没空。 I am busy at the moment. 我认为我行！ I think I can manage it. ② （指两人或以上） （用作主语） we （用作宾语） us （表所属关系） our 我厂/国/校/军 our factory/country/school/army 敌军被我全歼。 The enemy was annihilated by us. → 我方, 敌我矛盾 ③ （表泛指） [used together with 你 in parallel structures] anyone 大家你一言，我一语，献计献策。 They had a brainstorming session with anyone and everyone joining in. 市场里你来我往非常热闹。 The market is bustling with people coming and going. → 尔虞我诈, 你死我活 ④ （指自我） self → 忘我, 自我"

set outbound to ""

set cmd to "perl -we \"print <> =~ /([\\w+]..)/\" <<< "

set outbound to (do shell scriptcmd & inbound'squoted form)

display dialogoutbound

Thanks for the inspiration for a service, and for showing how to put a reference to another workflow in an AppleScript loop via 'do shell script'. I don't have your dictionary on my Mac (it doesn't seem to be on the US Mac App Store and I couldn't figure out where else to get it) but I do have freeware Xiao Cidian, which I've seldom used because it has lots of single characters but few combinations. That makes it quite well suited for this purpose because it's easier to get the pinyin out; just extract the first line of the definition (and filter out any "Not found" for when the script encounters punctuation or non-Chinese in the input.)

So this is now working here:

With this (placed in Workflows folder instead of Services, so it doesn't clutter up the menu; this also makes it available to the Run Workflow action in Automator in case you want that in the future):

So the result of selecting this: 第二十三回 横海郡柴进留宾　景阳冈武松打虎, then choosing Get Pinyin from the Services menu, and pasting in a text editor, is this:

Not bad, though the pinyin for some characters is not "context-aware."

The the script that works with xiaocidian is below. I have the output going to the clipboard but of course it could go elsewhere as well.

SG

on run {input}

set pinyin to ""

set inbound to input as text

repeat with c in inbound's characters

set def to (do shell script "automator -i " & quoted form of c & " ~/Library/Workflows/ch-to-py.workflow")

set def to def'scharacters 1 thru -2 as string-- strip "" and trailing return

set py to def'sparagraph 1 as string--get first line of xiaocidian.com definition -- the pinyin

try

if (py's characters 1 thru 2 as string) is "No" then

set pinyin to pinyin & c-- punctuation, etc. - "Not found"

else

set pinyin to pinyin & py

end if

on error

set pinyin to pinyin & py-- one-letter pinyin

end try

end repeat

set outbound to inbound & return & pinyin

set the clipboard tooutbound

-- return outbound

end run

I used the identical multi-lingual text that you originally posted for my testing. Not certain how you got the additional output from my AppleScript code. 😉

You can tell Perl to return the captured item as UTF-8, but when that change is made, wŏ becomes 我. For the same reason, it is why I did not set the outbound variable as text (or «class utf8»). Here is the Perl change that returns 我.

-- perl -we "print <> =~ /([\w+]..)/u" <<<

set cmd to "perl -we \"print <> =~ /([\\w+]..)/u\" <<< "

If you display the contents of /etc/paths, you will see that these are default paths already set up for you, and the processes that you launch.

Here is how you would incorporate the locale change. If you omitted the trailing 'u' on the Perl Regex, this would return wŏ.

set cmd to "export LC_CTYPE=UTF-8;perl -we \"print <> =~ /([\\w+]..)/u\" <<< "

The regex works here on the one sample multi-lingual text. Haven't figured out how to install dictionaries on Github page (use DictUnifier.app somehow?) so only have your one sample definition to work from. But it turns out that this returns the pinyin from your sample definition.

set py to input's paragraph 1's word 2

A rare case of AppleScript being more concise than a shell script! 🙂 (Can't speak for performance, though).

set input to "我 wǒ
①代说话人称自己。
我认识你 | 他是我的老师
②代称自己的一方，相当于“我们”。
我国 | 我军 | 我校 | 我厂 | 敌我双方
③代用于“你” “我”对举，泛指许多人。
你来我往 | 你一言，我一语。
④代自己。
自我介绍 | 忘我工作"

set py to input's paragraph 1's word 2

Instead of calling another workflow to look up a word it seems it *may* be possible to use open location "dict://...". May or may not be more efficient, and not sure whether there is a way to specify the dictionary.

SG

The Apple Dictionary app has a Simplified Chinese option in its preferences. One can click, and then drag this selection to the very top of that window so that it becomes the default dictionary. Previously, I had written an application in Python that uses Apple's Dictionary Services. Provide a native word on the command-line, and it returns the local dictionary entry. When I supplied 我 to the application, I got the following in return:

odin: ~$ pydict.py 我

我 wǒ①代说话人称自己。我认识你 | 他是我的老师②代称自己的一方，相当于“我们”。我国 | 我军 | 我校 | 我厂 | 敌我双方③代用于“你” “我”对举，泛指许多人。你来我往 | 你一言，我一语。④代自己。自我介绍 | 忘我工作

Because Apple's Dictionary Services will search all selected dictionaries in Dictionary app's preferences for the provided word, preferences order isn't important. The following curl dictionary lookup gets no match.

curl -s dict://dict.org/d: 我