Using OR in grep.

Hi.
i am having some trouble locating a way to do what I want.
I want to tail and follow a log file, which is easy, just tail -f /somelogfile.
Then rather than see the whole lot I want to pick bits out, for example if it was a BB2 and I want to just see posts I could do:
tail -f /some-log-file | grep some-entry
would produce what I want. However, what I want to see is lines that contain either some-entry, or some-other-entry or another-entry
So that I can pick out several types of interesting traffic with out having to have dozens of screens open.
Any one help? The man file implies I can do:
tail -f /some-log-file | grep some-entry |; some-other-entry |; a-third-entry
but this falls down with this bash error:
-bash: syntax error near unexpected token `;'

lanrat's option seems to work well, but what I would like to do now is to pipe the output to another logfile so I tried something like this:
tail -f /some_log_file | grep "first_expression|second_expression|third_expression" > /some_other_logfile
but tailing that shows nothing at all.
Any thoughts?

Similar Messages

How to go to footnote number automatically using Find or GREP

Hello,
I have been looking for how to search within footnotes. I have a Paragraph Style for my footnotes and can easily search anything within the citation but it won't take me straight to the number because I need to know what page it is at.
For instance the client is reading off his Microsoft Word doc and wants to edit the footnote no. 55. I don't know which page this footnote is at in the Indesign file, unless I just use the basic ^F with the "Find/Change" and manually go through it from no.1 until I reach no.55.
Isn't there a faster way to do this?
Thanks a bunch!

Here is a tiny Javascript that prompts you for a footnote number.
Caveats: it only works with numbered footnotes, not with "a", "b", "c" or "*", "**", "***" or "i", "ii", "iii" or any of the other numbering styles. It simply enumerates the current footnotes starting from '1' -- even if you have it start at another number, or have it restart numbering on every new page or section.
In addition, since every story (continuous thread of text) contains its own set of footnotes, you have to click the text cursor into your current story.
var notenum = prompt ('Footnote', '1-'+String(app.selection[0].parentStory.footnotes.length));
if (notenum && notenum.match(/^\d+$/))
app.select(app.selection[0].parentStory.footnotes.item(Number(notenum)-1).insertionPoints[0]);
app.layoutWindows[0].zoomPercentage = app.layoutWindows[0].zoomPercentage;
(Edit: Added the allowed range in the prompt)

Grep command in Solaris 10

Hi all,
I linux i can use this command
grep -C 3 err /u01/oracle/admin/ORCL/bdump/alert_ORCL.log
How can i do this in Solaris 10?
Thanks in advance!
Dan.

This work perfectly!
-bash-3.00$ /usr/sfw/bin/ggrep -C 3 err /u01/oracle/admin/ORCL/bdump/alert_ORCL.log
Thank you all!
Dân
Edited by: Dan on Jan 9, 2012 10:36 PM

How can I apply a GREP style to a text variable?

Hello everybody,
I have a question concerning GREP styles inside Paragraph styles.
1. I've created a text variable to generate a recurring title on the upper side of the page based on the main title paragraph style;
2. The recurring title is in Adobe Garamond Small Caps, all letters in lower case, and it is formatted with a paragraph style sheet in the master page;
3. I want to create a GREP style for the recurring title, according to which every time that in the recurring title appear an apostrophe or the double quotes, they are automatically lowered 2pt on the baseline
(I already created the character style sheet that lowers letters of 2pt).
What I need is the correct GREP formula to automatically apply the character style sheet to apostrophes and double quotes, in the line of text generated by the text variable...
Thanks for your help
p.

Hi,
As I said, using Power Headers is the best way to do it.
As Power Headers treats the header as "live text", you can use a simple grep style inserted in the header para style:
… to obtain:
For the sample, I use a char style named "-2pts" with Shift -5 pts and Green color to show you the place of ' and ".
Don't forget that, even Power Headers treats the header as "live text", you only have to update Power Headers to make an update of the headers!
Even I use in another cases Tomaxxi's [JS] and it's a good way to treat the question, Jean-Claude Tremblay's solution is less interesting because the variable used is converted in text. If the variable text content changes, it's more complicated to manage the update!

How do I use the commands to start and restart the jmqbroker normally?

How do I use the commands to start and restart the jmqbroker normally?
Because I have tried to use "jmqcmd shutdown bkr" and then tried to type
"jmqbroker" to start up. My screen hangs up and seems that the process
hangs up. The message is
[28/Jun/2002:18:29:07 GMT+08:00] [B1060]: Loading persistent data...
[28/Jun/2002:18:29:08 GMT+08:00] [B1039]: Broker "jmqbroker" ready.
The prompt is hanged until I press CTRL-C.
Then I got the following message:
[28/Jun/2002:18:32:07 GMT+08:00] [B1047]: Shutting down broker...
[28/Jun/2002:18:32:07 GMT+08:00] [B1077]: Broadcast good-bye to all
connections ....
[28/Jun/2002:18:32:07 GMT+08:00] [B1078] Flushing good-bye messages
[28/Jun/2002:18:32:07 GMT+08:00] [B1056] Cleaning up persistent
store...
[28/Jun/2002:18:32:07 GMT+08:00] proceessing 0 messages and cleaning
up 0 files...
[28/Jun/2002:18:32:07 GMT+08:00] [B1063] Done
[28/Jun/2002:18:32:07 GMT+08:00] [B1048]: Shutdown of broker complete.
Then, I use jmqadmin to invoke the console, I can start the broker
again, it gives me
Error encountered while connecting to the broker: "MyBroker":
Broker Host: 'localhost'
Primary Port: '7676'
[C4003]: Error occurred on connection creation, - caught
javax.jms.JMSException
Please verify that there is a broker running on the specified host and
port.
I used ps -def | grep jmq there is no process anymore.
Is it a bug? Would somebody have any idea on what went wrong?

'jmqbroker' is the command to start a broker.
When you type 'jmqbroker' in a command window,
unless you put it in the background, the shell
prompt won't return until the broker is shutdown.
So when you see 'Broker "jmqbroker" ready.', it tells
you that the broker is running and is ready.
When you CTRL-C the process, it has the same effect
as using jmqcmd to shutdown the broker. And so after that,
there's no jmq processes running since you have just
shutdown the broker.
Remember that the jmqcmd/jmqadmin 'restart' command is to
restart an already running broker. jmcmd/jmqadmin does
not support starting a new broker instance.
If you are using csh, end the jmqcmd command with an ampersand ("&")
to have the broker running in the background. An alternative is to
stop the process with Ctrl-Z then input the command "bg" to have that
process resumed in the background.
The problem with not being able to start a broker using the
administration console is a documented limitation of the product that
Yvonne also mentioned. Check the Administrator's Guide for that.

[CS3] GREP-Problem with Anchored Objects

Hi,
I've the following problem:
Lines beginning with a, b, c etc. should be formatted equally, so I use the following GREP:
Search:
^([a-z])(\.??)([ \t])
Replace:
$1.\t
Now there are some abc lines, which begin with an anchored object, so I tried:
Search:
^(~a*?[a-z])(\.??)([ \t])
Replace:
$1.\t
But with this, the anchored object is deleted. Is this a bug or am I doing something wrong?
Thanks
Tobias

That's a known bug -- Dave Saunders discovered it, to his surprise, I might add. There seems to be no reason for it.

Not able to stop database using srvctl

When i stop database using srvctl from node1.. "srvctl stop database -d mosa" it does not stop. I tried "srvctl stop database -d mosa -o immediate" but to no luck. Could anybody help me on this issue.
OS = HP-UX
Database = Oracle 9i
Instance1 = mosa1
Instance2 = mosa2
node1 = hp1
node2 = hp2
DB Size = 50GB
I have opened another session to check srvctl status
$srvctl status database -d mosa
Instance mosa1 is not running on node hp_data1
Instance mosa2 is running on node hp_data2
NOTE:- BEfore i stop database, i had stopped listener.
Regards,
Vikhar

Thank you , issue is solved.
After i give below given command, i checked the status by
$srvctl status database -d dbname
$Instance mosa1 is not running in node1
$Instance mosa1 is running in node2
It means PROBLEM with instance2
I have got following error message in alert_log file
attempting to shut down the database and the database hangs at $srvctl stop database -d dbname. The alert log contains the following message:
SHUTDOWN: Waiting for active calls to complete
There are no other error messages in the alert log.
+Solution Description:
=====================
Caveat: This note is intended for Unix platforms primarily, but we should
address this issue on Windows platforms as well. On Windows, stopping the
service is a valid workaround for these errors.
Locate and kill any client connections to the database at the Unix level, as
follows:
1. Locate any client connections to the database using ps, and grep for any
processes belonging to this .
Example: ps -ef | grep V733
2. Look for processes that include a 'Local=No' designation.
Example: osupport 6235 1 0 Nov 24 0:01 oracleV733 (LOCAL=NO)
3. Kill the Unix process(es) with the 'Local=No' designation.
Example: Kill -9 6235
Explanation:
============
The database is waiting for pmon to clean up processes, but pmon is unable to
clean them. The client connections to the server are causing the shutdown
immediate or normal to hang. Killing them allows pmon to clean up and release
the associated Oracle processes and resources+
What resources are we talking about?
1) Any non committed transactions must be rolled back
2) Any temporary space (sort segments / lobs / session temporary tables) must be freed
3) The session itself and any associated memory consumed by the session.
4) Internal locks / enqueues must be cleaned up
Often Oracle (SMON or PMON depending on whether Shared Server is used) will wait for the OS to terminate the process(es) associated with the session. I the OS never returns, or fails to terminate them, then the instance shutdown will hang with this message (Shutdown Waiting for Active Calls to Complete) Other means exist to achieve a quick shutdown.
Thank you so much for your efforts.
Edited by: VIKHARAHMED on Feb 27, 2011 10:19 PM
Edited by: VIKHARAHMED on Feb 27, 2011 10:25 PM

Why doesn't Grep Styles with pos lookbehind work?

Hi,
I've posted another thread but I suppose I expected too much. I'll try to siplify this.
I have text where I want to catch underlined text:
Ponny, - Cob…….499,-
125, - 135, - 145, - 155…….399,-
140, - 150, - 160, - 170……..99,-
140, - 150………99,-
This paragraph grep style works for the above: ^\w+,\s-\s
However, when I want to use positive look-behind to just catch the " - " it stops working.
I have tried the following (none of them catches anything at all):
^(?<=\w+,)\s-\s
(?<=^\w+,)\s-\s
(?<=\w+,)\s-\s
Is there a limitation as to what syntax I can use in the lookbehind?
Pointers anyone?
/K

Hi Eugene,
Lookaheads and Lookbehinds don't work when using the repeat codes, like "+" or "*" or "?" etc.
I guessed there might be a limitation as to what works.
Eugene Tyson wrote:
Can you search for
\s-\s
And only search for underlined text in the
Find Format
area
No, sorry. The underline was olnly to indicate what I was trying to grep...
I don't want to use find-replace. I want to use the paragraph grep style.
/K

Can I use Boolean parameters and 'is empty ' in smart groups & mail rules?

How do you do the following smartgroup in Mail? State is MA or NH or ME and Email is not 'empty'? I also would like to make the following rule in Mail: If FROM is 'empty' and Subject is 'empty' then move to Trash. In other words is it possible to use invisible charakters <blank> or <empty> and boolean parameters in search fields. Any hits will be appreciated. ---Stefan

I have the same question, plus a couple more. Most of the spam that I get follows one of five simple patterns:
1. It contains one or more empty fields (to, from, subject, or message body),
2. It sometimes lacks a "from" or "subject" field entirely.
3. Mail recipients are sequential (e.g. [email protected], [email protected], [email protected]).
4. Total message size is 0.7 KB or less. (Due to empty fields.)
5. It contains a gif or jpeg touting whatever product the spammer is flogging.
It seems like it would be trivial to incorporate filter rules to take care of empty or non-existent fields (categories 1 and 2). But I can't see a way to do that in Mail currently. Hopefully Apple is working on this? Or maybe there's a third-party add-on that can address this?
Also it would be easy to take care of #3 with the use of simple grep expressions. But I don't think Mail can do this yet. (Again, any third-party solutions out there?)
For #4, obviously a rule that could take into account message size would help deal with those emails.
And for #5, it's more problematic, but since there are programs out there that can read graphics, hence the need for captcha "encryption" on web sites, why not use such an app as a Mail extension? It would be great if it could interoperate with Mail's rules, so you could set up something like "if message contains a jpeg or gif and if the image contains 'penile enlargement' then delete/put in junk folder."
Currently I put all messages with attachments (that haven't already been approved by my whitelist) in a "limbo" folder for later review. But I have to use a minor kluge to do this. Ideally I'd just say "if message has an attachment" or "if message attachment is not empty," but those options aren't available so I have to say "if name of attachment contains '.'" Since pretty much all files have a "." in them, it seems to work fine, but strictly speaking it's not as elegant as I'd like it to be.
iMac G5 iSight, Powerbook Ti Mac OS X (10.4.5)
PowerBook G4 (Ti) Mac OS X (10.4.1)
iMac G5 iSight, Powerbook Ti Mac OS X (10.4.5)

GREP Question: Style em spaces that precede only certain triggers within a paragraph

Hello,
Up front:
I do not know GREP well enough to begin to tackle this. And as usual, I'm in a pinch!
Also, if you're the visual type, just look at the second to the last paragraph at the bottom, and that may be enough to help you devise the GREP style. Otherwise, bear with me, I have to think this through...
I have a list that spans pages of events. The events are formatted as follows:
☐Box City EventNameSpecific Month EndPara (this is a single paragraph in a story of a hundred such paragraphs)
On paper, each event will need to be preceded with a filled specific-colored box (☐). I would like this box to be made from a strikethrough applied to an em space using grep, since that is very easy to create using a character style and only requres one setting (the thickness of the strikethrough).
There is a legend at the bottom of the page, detailing about 14 types of events. Some of the different types of events will require the same colored box. But in all I have 7 different colors of boxes in the legend.
I'm looking to create...
7 character styles, one for each color (consider this done already)
1 paragraph style (essentially done, but missing the GREP)
7 GREP styles (or more as necessary) specifically to apply only to the em space that precede specific sets of text strings
If I need, or if it is simpler to create 14 GREP styles within the paragraph style, that's fine. But I understand we may be able to do something like this... (Target phrase 1 | Target phrase 2 | Target phrase 3) in between parenthesis to identify multiple possible triggers within the same rule.
I'm fine whether it's 7 or 14 rules in there, I just want to be able to update them over the course of time as necessary - but it won't be for another season at least if I do.
I don't know GREP well enough to piece together what I need. Vaguely familiar with look aheads and look behinds. I do understand it may be easier if the em space is in between other characters, so in my example below I've inserted hairspaces. But ultimately I want the em space to be flush or nearly flush to the left of the frame.
For my example below:
I only want to apply character styles to em dashes based on the strings in bold.
The strings of text that will trigger the GREP will vary in number of words and may or may not contain a dash.
I want to ignore Citynames regardless of number of words... Chicago vs. San Francisco vs. Vancouver, BC for example.
I want to ignore MonthofYear which appears after the second tab.
The tabs will have set stops as a regular part of the paragraph style. I'm including them in my example below in case it helps to visualize possible anchors to use in GREP.
I want the style to be easy enough to modify, for example if I want to simply add an (EEEE New Eventname) to the list of possible options within a rule that triggers a specific character style. See below should make it clearer.
So here's what I'm thinking...
Within the Paragraph Style, the first two examples of GREP Styles should be able to respond to the following criteria:
Rule 1: If the paragraph contains any of the following, then apply Char Style 1 to the em dashes in those paragraphs. Use a unique GREP expression for each phrase below where each uses Char Style 1 or use a single GREP expression to capture all three possible triggers:
ABC BB Invitational
ABC Xxxxxxxxxxxxxx
ABC MM-Xxxxxxxxxxxxxxx
Rule 2: If the paragraph contains any of the following, then apply Char Style 2 … and so on
Global ABC Xxxxxxxxxxxxxxx
Global WXYZ Xxxxxxxxxxxxxxx
I cannot count on “ABC” or “Global” being the trigger for the style, if you know what I mean. I need the entire phrase (ABC BB Invitational) to be the trigger… if it exist in its entirety, then apply the style to the preceding em dash in that paragraph. This way if there are any mispellings or if we launch a new event type which ends up flowing in to my document I will know it.
hairspace emspace tabspace CitynameOneWord ABC BB Invitational tabspace MonthofYear
hairspace emspace tabspace Cityname TwoWrds ABC BB Invitational tabspace MonthofYear
hairspace emspace tabspace Cityname MultiWrds ABC BB Invitational tabspace MonthofYear
hairspace emspace tabspace Cityname TwoWrds ABC MM-Xxxxxxxxxxxxxx tabspace MonthofYear
hairspace emspace tabspace Cityname MultiWrds ABC Xxxxxxxxxxxxxx tabspace MonthofYear
hairspace emspace tabspace Cityname TwoWrds Global ABC Xxxxxxxxxxxxxxx tabspace MonthofYear
hairspace emspace tabspace CitynameOneWord Global WXYZ Xxxxxxxxxxxxxxx tabspace MonthofYear
hairspace emspace tabspace CitynameOneWord Global Special Xxxxxxxxxxxxxxx tabspace MonthofYear
hairspace emspace tabspace CitynameOneWord Globl Special WRONG SPELLING tabspace MonthofYear
I hope this makes sense and isn't too unnecessarily redundant. Time for bed. Fingers crossed someone will post at least one GREP string, so I have some magic code for tomorrow morning! Ideally, it would be great if you would include a brief explanation of waht the string is doing, but minimally, please do use one of my text stings above, so I know what to mess with and what not to. ;-)
<says prayer>
Thanks!
Rick

Well, for a moment I thought I had an idea of how to do this with GREP styles, but I realized it wouldn't work, and I'm not sure it's possible unless you are able to make a separate style for every possible string between the em space and the trigger text. You want to use the lookahead because you want to find that text, but you don't want to select it or apply the style to it, but a lookahead can't handle the kind of variability on string length that you need to work based on just the bolded text at the end.
It would actually be easier to use paragraph styles based on the bolded text that include a nested character style, I think. That could be handled by find change, but I think the best thing is probably to move this over to scripting.
Rather than just move the discussion, though, I'm going to ask you to cross post so there are two threads on the chance that someone else who doens't visit scripting might have another idea for you here.
Here's a link to scripting: InDesign Scripting

Attempting to create a GREP Style

I'm trying to make a GREP style for one of my paragraph styles, that will take the text from the beginning of a line until a colon and set a character style to it to bold that part.
Example:
Step 1: Take bread
Step 2: Take peanut butter
Step 3: Spread peanut butter on bread
I don't think I've been getting the syntax right. I've put ^\:~h and ^:~h into the 'to text' box, thinking this meant "Beginning of line until the character ':' and end" but neither of those have given me the results I want. Thanks in advance for the help!

Prismatus wrote:
Your response worked for me. I had tried using a wildcard before the ones I posted, but only put the . in, assuming it would cover more than just one character. I assume the +? covers all characters until the colon shows up, yes?
No, it's slightly more complicated than that (sorry!). By default, GREP is Greedy -- that means, if you use this
^.+:
GREP will think that the Any Character wildcard may be repeated as much as possible (that's the '+') before it needs to match the colon. What this means is that it will work just as you expected for
Step 1: This is a single line.
but will go out of its mind with this
Step 2: what will happen now? Well, contrary to what you were expecting, the entire line will be marked bold, all because everything up to the very last : will be matched!
The bold bits accurately shows what happens! Another example would be this:
\d+
which for a string of "123" will not match just the first digit, then the second, then the third, but all of them in a single long go. By default, GREP will grab as much as it possibly can.
Adding a question mark behind the "Repeat Me" character reverts this behavior to Non-Greedy behavior, and as such GREP will match as little as humanly possible:
\d+?
will then match just the "1" in "123".
Prismatus wrote:
I also like the be literal part, but I was wondering what you would have instead of \d if you had more than nine steps, or if you had substeps (1a, 1b, etc.).
That's just a case of adding more specifiers. To match one or more digits, you need this:
^Step \d+:
The '+', again, will allow a repetition of the "digit" code. In this case you don't have to add a question mark, because there is no way this could run out of control; first, it will only match digits, and second, these digits must be followed by a colon.
And if you may or may not have a single lowercase character following the digit (but still before the colon), you'd use the Any Lowercase Character code and the Zero or Once repetition specifier:
^Step \d+\l?:
(that's a lowercase ell.) You see the question mark meaning Something Entirely Different here? It's only a Non-Greedy marker when immediately following another repetition code, one of these
* (zero or more)
+ (once or more)
? (zero or once)
{4,8} (or any other set of numbers -- this is at least 4 and at most 8 times)
Uh, by the way, that makes this
abc??
a valid GREP. It will match "ab" or "abc", and then always select the shortest possible match of these two, which is then the "ab" one. ... Uh. I'm pretty sure this may be useful, some time in the future.

Find/Change - GREP vs. TEXT

Today, thanks to this amazing forum, I discovered that I had been using "Find/Change" under TEXT, instead of GREP.
Using "Find/Change" as GREP for finding & replacing seems to be MUCH MORE POWERFUL than "TEXT."
What are the disadvantages of using "Find/Change" GREP ? I notice it cannot do "case sensitive" — any other cons?

GREP searches can be either case insensitive on or off
case insensitive on: (?i)
Case insensitive off (?-i)
Example:
Search for either "case" or "Case" (?i)case found on 1st, 2nd and 3rd line
Search for just "Case" (?-i)Case found on 3rd line only

GREP Styles and Changing Ligatures

Hello,
I have been using InDesign since it came out, and have great results. However, I am brand new to GREP, and was wondering if a script could be made/written(?) by me (a person who does not understand script), to do the following:
I have tons of copy, regularly, with a number of different ligatures in it. One (of the many, the fi ligature) causes the client to cringe. Is it possible to change this ligature by - turning off - the fi ligature only? Because I am out of the old school, I appreciate ligatures, and tight kerning.
I realize, I could just turn off ligatures, but, would rather not.
Thank you,
Lain Kennedy

A character style whose only property is no ligature, used as a GREP nested paragraph style, is certainly the best approach. However, if the ligature appears in many different paragraph styles, it might be possible to change them all in one operation by creating a paragraph style that duplicates (not based on) [No Style] - that is with no properties set - and create a GREP nested style for it that applies the no ligature property to fi. Then set each paragraph style to be based on this no-fi-ligature paragraph style.
I haven't tested this to see if this really works. Character styles are supposed to supersede paragraph styles, but perhaps not in this scenario.
HTH
Regards,
Peter Gold
KnowHow ProServices

Collect Thread Dump using Script and direct to separate log file

Dear Legends,
I request you to provide me some useful guide lines to collect Thread Dump using script in linux and redirect the output to a separate log file, not the STDOUT file. If it is redirecting to STDOUT then how to extract only the thread stack to a separate log file.
i used
ps -ef | grep java
kill -3 <pid> >> ss1_td.log
but it is not providing the thread stack to the log file.
Anykind of help would be appreciated much.
Regards,
Karthik

Hi Ram,
Really a great and useful script for me at the right time. It solved my requirement and i made some changes to my requirement and the script it working fine to collect the thread dump and even i have checked the thread dump with SAMURAI works fine.
Helpful site also refer sites like these which contains more valuable things for weblogic. We will meet in the forums again.
Note: also please update several links available in the site like UNIX scripts, WLST, Python etc.
Regards,
Karthik

Characters chopped off when using 'ps -ef'

Hello all:
When I monitor a java process, I use this commond:
ps -ef | grep MyJavaServiceit appears that MyJavaService has a long jar file list prepended as the classpath, the above command didn't show all the file names on the console - it's chopped in the middle.
Is there a limit of the numbers of the characters shown when using ps -ef | grep someProcess ?
Thanks,
Johnny

ps -ef | grep MyJavaServiceThere is a system limit on line length, check with:
# /usr/bin/getconf LINE_MAX
2048LINE_MAX: maximum number of characters that a utility can accept as an input line (either from the standard input or a text file) when the utility takes text files as input. This number includes the trailing newline.
Are you piping this through a pager such as less(1)? The ${LESS} environment variable can control whether long lines wrap or truncate.
What terminal emulator are you using? These can also be set to wrap or truncate long lines.

Using OR in grep.

Similar Messages

Maybe you are looking for