Generate best matching regex/rule from string set

does there exist a tool which:
given a set of strings, will parse those strings and output a the simplest possible regex which matches those strings
or
does there exist a tool which, given a regex, can output a simplified(bonus points for optimal) version of the same regex
?

i should rephrase, simplest regex that matches a set
of strings, and nothing more. i can make some for the
set: "doggy, dogma, man" -> doggy|dogma|man but if
my set of strings is very large it might be better to
try to write the regex like this: dog(gy|ma)|man
so, the real problem is that i have several sets of
500+ strings, i need to write a short/efficient regex
to match them.What do you need it for?

Similar Messages

While trying to sync my calendar to my new mac, I lost all my calendsr events on my iPhone. Best guess is I calends't set up icloud on my phone first and my computer delted them. Is there any way to recover them from iCloud?

While trying to sync my calendar to my new mac, I lost all my calendar events on my iPhone. Best guess is I didn't set up icloud on my phone first and my computer deleted them. Is there any way to recover my calendar from iCloud?

Your music will only be where you put it.
Did you copy all of your music from your backup copy of your old computer to your new one?

Regex: How do I match only when a string is not entirely numeric?

Hi,
I'm using java regular expression pattern matching. I want to receive a successful pattern-match ONLY when a string is NOT comprised entirely of numeric digits [0-9].
So, for example, when pattern-matching on the following 3 strings,
"abcdef" would return match = true and
"abc123" would return match = true and
"123456" would return match = falseI have tried using the regex pattern [^0-9] but this only seems to work for strings that are one character long. The pattern [^0-9]+ didn't work either.
What I need is the equivalent pattern to [^\d+] (which is an invalid pattern).
Please note that I have to solve this one using regular expressions, even though it would be trivial to solve using straight java code!
Any help much appreciated.

What I need is the equivalent pattern to
[^\d+] (which is an invalid pattern).Well as its an invalid pattern, how exactly do you expect us to know what you mean by it?!?
Based on your textual description of what you want, and on the basis that you seem not to want to negate theopposite, how's this (I assume we're talking core Java regex's?):
myString.matches(".*[^\\d].*");

Best way to remove unwanted signs from string

Which is the best way to purify a string from possible dangerous signs? I want to remove all characters that is not a-z or 0-9? how would I go about?

read the string character by character, and check to
see if they're characters you want to remove or not,
by comparing them to the ascii values of characters
you permit. I've used this before, but I can't
remember what exactly it permits, and I'm too lazy to
look up the ascii table :) But this sort of thing
works.
int c = in.read();
while(c!=-1){
if((c<33)||(c<58&&c>44)){
out.write(c);
I thought about that too. But the values in the ASCII table... arn't they a little volatile to use? I mean, will my application work on other systems or systems with other languages?

What is the Scan from string pattern for "match everything" ?

Hello,
Using Scan from string for a while, I know that %s only matches string up to a whitespace. And I also thought %[^] would match everything including whitespaces. But it turned out that it would stop at the closing square brace.
So, what is the real scan pattern for match everything including whitespaces ?

What do you want the Scan From String to end on? Or are you just grabbing the rest of the string? If that is your scenario, then just use the "remaining string" output. It might help if you give a full example of a normal input string and EVERYTHING you want as an output.

Urgent pl. I want to extract float from string (%f %s) or (%s %f) separate

I want to extract float from string (%f %s) or (%s %f) separated by tab, whitespace etc., Since I am using jdk1.3.1 I can't use regular expression can anybody suggest a simple (one or two or few line code - compact) to get the float in (sign+/-)#.#### format ignoring other characters ?
I tried:
e.g.,
String d="4.000 [tab]4";
source string resulted from "SUBSTRING query of mySQL" can be of :
" -4.543 XYZ ",
" XYZ -4.546 ",
" xx-yy 6.58 3 ",
"6.0 xxx yyy zzz",
the expected results for float from the string should be:
-4.543
-4.546
6.580
6.000
If String containing data separated by comma:
e.g.,
"4.120     1     ,
AAXXFE     ,4.206     1     ,
     4.000     1,
     4.201     1,
     4.189     1,
     4.204     1,
S     DDERSF-RSA"
The result should be: "4.120,4.206,4.000,4.201,4.189,4.204,0.000"
The string is created by:
ResultSet rs1 = stmt.executeQuery(S1);
while (rs1.next()) {
String d = rs1.getString("SUBSTRING(FIELD1,LOCATE(\""+s2[i]+"\",FIELD1)+"+ k_st + ","+ length+")");
System.out.print(d+",");
May be I don't know how to use the following:
float f1 = Float.parseFloat(d);
===
String pattern = "###.###";
//float value = -2100.578f;
DecimalFormat myFormatter = new DecimalFormat(pattern);
String output = myFormatter.format(d);
System.out.println(d+" " + pattern + " " + output);
===

Hi
You made the best choice. JRegex is closer to java.util.regex.* and fine works with excellent performance in any JVM version so you can use examples/tutorials from both JRegex and Java.Sun.Com to learn more about regular expression.
About the regular expression in my last post, there are two capturing groups: the first for float numbers in IEEE format specification (also see Java Language Specification) and the second for any sequence of chars, groups using with any amount of blank spaces (spaces, tabs, etc :: see regular expression definition) as delimiter. The anchors "^" and "$" are being used to exactly match pattern in target string and you can relax this constraint as needed.
How to use JRegex? no secrets!
import jregex.*;
// a float number and string with any blank delimiter
String patternString ="^([+-]?(?:\\d+(?:\\.\\d*)?|\\.\\d+)(?:[eE][+-]?\\d+)?)\s+(.*)$";
Pattern pattern = new Pattern(patternString); // different from Sun regex package syntax
// below is identical in both packages
Matcher m = pattern.matcher(anyTargetString);
if (m.matches()) {
Double d = Double.parseDouble( m.group(1) );
String scratch = m.group(2);
// your stuff goes here
..Remember to put JRegex jarfile in classpath!
Regards and Success.

Need to import Rules from 10.3.9 to 10.4

I just made a clone of my existing 10.3.9 OS and all folders to a FireWire external hard disc, which is bootable. In fact, I am using Safari on the old OS in order to submit this post, since I am finding the new OS a bit of a nuisance to set up.
So you know, I did a clean install of OX 10.4. and did not elect to import apps, or any other personal files. I did however, import the network settings. It irks me to no end that Apple has not made the few categories you can check at this point in the installation a wider variety or hierarchical (so you can choose which apps to bring over, not just all of them).
As for Mail, the new version of the software made importing my existing accounts a laborious affair, and there appears to be no way to import Rules. I've got a lot of them, and relied heavily on these to make my emailing experiences tolerable.
PLEASE, how do I get the newer, "improved" Mail in 10.4 to import my Rules from the older version of Mail in 10.3.9?
Thanks in advance for the help.

"Mail doesn't import rules", you say...
Boy, THAT makes a lot of sense, doesn't it? Since the program evidently still uses Rules...(eyes roll)
Thanks for the suggestion, by the way; I tried the .plist copy-over right away and discovered the "new improved" Mail would not recognize the prefs...
Y'know...since the "new" Mail program still offers and utilizes Rules (apparently the same way the older version did), not providing a way for users to import settings like this (that can take tons of time to replicate) seems purdy darn'd inconsiderate.
Rules are the one significant way I have found to make email efficient -- not just Apple Mail, but any mail program. Call them what you like: rules, filters, etc. They are indispensable; evidently developers thought they were still worth offering, but lifted their legs on the time and effort end users put into creating these settings. Either that, or the program was made willfully ignorant as part of some great plot to hamper everyone's productivity. No, just mine.
I apologize for the facetiousness, but I'm really perplexed (and angry) that developers would either overlook something basic like this or make the latest version of the program so willfully ignorant.
At least my old Mail program still resides on my bootable HD (the best thing I ever did with my Mac was cloning the old OS to another drive so I keep going back to it until I wade through these ridiculous set-backs).
My inner child will love this: now I get to take physical Pen And Paper notes of my old rules so I can ultimately recreate them in the new and improved but willfully ignorant Mail program. I have to laugh... Pen And Paper to solve a computer problem. Gotta love it...
...maybe it'll function as some kind of weird Zen exercise? I'm hopeful!
Thanks for the input once again, and Peace to You (...but not to the folks who developed the latest version of Mail).

How to create business rules from Jdeveloper 10.1.3.1.0

Dear Experts
Can I able to create Oracle Business rules from My JDeveloper 10.1.3.1.0. Using this link They are creating repositories, rule sets for their BPEL process, But in my version I cant found <BPEL components> under component palette. Why these features are missing, to get those additional features what I have to do.. ? Please suggest your best for me.
Also Is it possible to build business rules from java sdk & from rule author..? If so means what is the difference, which one is having advantage.
Thanks,
Rajesh

http://download.oracle.com/docs/cd/B31017_01/integrate.1013/b28981/decision.htm

How do i return the count of matching characters in 2 string

hi guys,
is there any comparision function in oracle that return the number of matching character between 2 string ?
example - 'josephine' and 'johnny'
return 3 ('j','o' and 'n')
please advice,
Best Regards,
Noob

Hi,
Type creation:
CREATE TYPE t_character AS OBJECT
   v_character varchar2(1)
CREATE TYPE characterList AS TABLE OF t_character;
/Function creation:
CREATE OR REPLACE FUNCTION fnc_compare_string(p_string_1 IN VARCHAR2,
                                              p_string_2 IN VARCHAR2) RETURN VARCHAR2 IS
   RESULT VARCHAR2(4000);
   v_list1 characterList := characterList();
   v_list2 characterList := characterList();
   v_first BOOLEAN := TRUE;
   c_separator CONSTANT VARCHAR2(1) := ',';
BEGIN
   FOR i IN 1 .. length(p_string_1) LOOP
      v_list1.EXTEND;
      v_list1(i) := t_character(substr(p_string_1,
                                       i,
                                       1));
   END LOOP;
   FOR i IN 1 .. length(p_string_2) LOOP
      v_list2.EXTEND;
      v_list2(i) := t_character(substr(p_string_2,
                                       i,
                                       1));
   END LOOP;
   FOR reg IN (SELECT DISTINCT *
                 FROM TABLE(CAST(v_list1 AS characterList))
               INTERSECT
               SELECT DISTINCT * FROM TABLE(CAST(v_list2 AS characterList))) LOOP
      IF v_first THEN
         v_first := FALSE;
         RESULT := RESULT || reg.v_character;
      ELSE
         RESULT := RESULT || c_separator || reg.v_character;
      END IF;
   END LOOP;
   RETURN(RESULT);
END fnc_compare_string;Test:
Connected to Oracle Database 10g Express Edition Release 10.2.0.1.0
Connected as hr
SQL> select fnc_compare_string('post', 'most') from dual;
FNC_COMPARE_STRING('POST','MOS
o,s,t
SQL> select fnc_compare_string('post', 'jhon') from dual;
FNC_COMPARE_STRING('POST','JHO
o
SQL> select fnc_compare_string('post', 'was') from dual;
FNC_COMPARE_STRING('POST','WAS
s
SQL> select fnc_compare_string('post', 'w') from dual;
FNC_COMPARE_STRING('POST','W')
SQL> Regards,

Migrate Business Rules from 9.x to 11.x

I am trying to migrate business rules from 9.21 to V11.1.1.3
I used to export the rules, then edit the XML for the correct server and location names, then import into the target server-all with 9.21
Now when I try the same process, I get this error when importing into V11.1.13
Detail:Error while importing object properties.
Detail:com.hyperion.hbr.beans.UserObjAccessFilter.setUserGroupName(java.lang.String)
I can use the migrate repository option just fine, but then I have to go in with a SQL script and update the location and server names in the relational tables. I prefer the export/import since I can do that without impacting users.
Any ideas on how to import the rules and resolve this error?
Thanks
Jeff

Hmm no luck with one rule that has no security assigned to it. and no luck with one project with no security.
Here is the xml for 1 rule that won't import:
<?xml version="1.0" encoding="UTF-8" ?>
- <HBRObjects>
- <Rules>
- <Rule>
<property class="java.lang.String" method="setBody" value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
<property class="java.lang.String" method="setMetadataReferences" value="" />
<property class="java.lang.String" method="setDefMetaDataRepo" value="Planning/balt-hypplan/CorpBud/Eeplan" />
<property class="java.lang.String" method="setBodyAsCalcScript" value="/*ESS_LOCALE English_UnitedStates.Latin1@Binary*/ SET AGGMISSG ON; SET UPDATECALC OFF; FIX (Budget, "1st Pass", &CurrBudYr, @LEVMBRS("Period", 0),@relative("Total Employees",0),@RELATIVE("Total Company",0)) CALC DIM ("Account"); "Employee Taxes" = "Employee Tax Rate Input"->"No Entity"->"No Employee" * "Total Salary"; ENDFIX CALC DIM ("Employee",Entity);" />
<property class="java.lang.String" method="setName" value="Corp-EE Benefits & Taxes Calc" />
<property class="long" method="setLastModified" value="1211738336600" />
<property class="java.lang.String" method="setVersion" value="" />
<property class="int" method="setId" value="220" />
<property class="java.lang.String" method="setDescription" value="" />
<property class="java.lang.String" method="setOwner" value="native://DN=cn=911,ou=People,dc=css,dc=hyperion,dc=com?USER" />
<property class="int" method="setObjectType" value="1" />
<property class="boolean" method="setEditable" value="false" />
<property class="boolean" method="setLocked" value="false" />
<property class="java.lang.String" method="setLockedBy" value="0" />
<property class="java.lang.String" method="setUpperName" value="CORP-EE BENEFITS & TAXES CALC" />
<property class="java.lang.String" method="setModifiedBy" value="native://DN=cn=911,ou=People,dc=css,dc=hyperion,dc=com?USER" />
<property class="java.lang.String" method="setOpenForEditingBy" value="0" />
<property class="boolean" method="setOpenForEditing" value="false" />
</Rule>
</Rules>
<Macros />
<Variables />
<Sequences />
<Projects />
<RuleMacroLinks />
<RuleVariableLinks />
<MacroMacroLinks />
<MacroVariableLinks />
<SequenceItemLinks />
<SeqPrompts />
<ProjectItemLinks />
- <UserObjAccessFilters>
- <UserObjAccessFilter>
<property class="int" method="setObjectID" value="220" />
<property class="int" method="setObjectTypeID" value="1" />
<property class="int" method="setLocationID" value="9" />
<property class="int" method="setTaskID" value="2" />
<property class="java.lang.String" method="setUserGroupID" value="0" />
<property class="int" method="setUserGroupTypeID" value="0" />
<property class="boolean" method="setSearchNoLoc" value="false" />
<property class="boolean" method="setSearchNoUser" value="false" />
<property class="boolean" method="setSearchNoObj" value="false" />
<property class="boolean" method="setSearchNoTask" value="false" />
<property class="java.lang.String" method="setUserGroupName" value="" />
</UserObjAccessFilter>
</UserObjAccessFilters>
- <Locations>
- <Location>
<property class="int" method="setCluster" value="0" />
<property class="java.lang.String" method="setLocation" value="Planning/balt-hypplan/CorpBud/Eeplan" />
<property class="int" method="setLocID" value="9" />
<property class="java.lang.String" method="setUpperLocation" value="PLANNING/BALT-HYPPLAN/CORPBUD/EEPLAN" />
</Location>
</Locations>
</HBRObjects>
I see the setUserGroupName property class and it is empty..

Error 116 at Unflatten from string.....

Hi I want to save and load the control values of
the tab pages I am getting the following error
"Error 116 occurred at Unflatten From String
Unflatten or byte stream read operation
failed due to corrupt, unexpected, or truncated data."
the attached is the VI can
anyone tell me what is wrong ….thank you
Attachments:
save n load example.vi ‏70 KB

All that you need to do is right-click on the Read from Text File and Write to Text File functions and uncheck "Convert EOL".
Other comments:
You should not hard-code the path inside the VI. Either place a front panel control to specify the path (perhaps with a default value set), or generate the path so that it is relative to the VI (such as being in the same directory as the VI).
You are not wiring all your errors through so some errors can be lost.
You do not need to have the VI open a reference to itself. If you delete the Open VI Reference function the code will still work.

ClassCastException when testing rules from WSNavigator - BRM

Im using SAP NetWeaver 7.2 (SP03 PAT0031)
I have created a rules flow which has 4 paths; when I test the paths, 3 return the expected results. My problem is with the final one. I have pinned down the error to a rulescript which executes multiple decision tables as well as rules. The error I'm getting is shown here:
Web service returned error. Fault Code: "(http://schemas.xmlsoap.org/soap/envelope/)Client" Fault String: "BusinessRules ERROR: com.sap.brms.qrules.network1.ActionException: java.lang.ClassCastException: class java.lang.Long:null incompatible with class com.sap.brms.qrules.util.Value:library:tcbrmsengine~lib @ com.sap.engine.boot.loader.ResourceMultiParentClassLoader @ 73c94b51@alive"
I'm not using any Long values in my xsd or anywhere else;
When I test the rule from the NWDS, it works fine, and gives me the expected result.
I'd appreciate it if anyone could shed some light on this issue.
Thanks.

Hi,
1. BPEL Ver 10.1.3.x
We have a process which accepts UserToken. I found out that this exception is thrown when soap:Header is added to the message.
<soap:Header>
<wsse:Security xmlns:wsse="http://schemas.xmlsoap.org/ws/2002/07/secext">
<wsse:UsernameToken xmlns:wsu="http://schemas.xmlsoap.org/ws/2002/07/utility">
<wsse:Username>user</wsse:Username>
<wsse:Password Type="wsse:PasswordText">pass</wsse:Password>
</wsse:UsernameToken>
</wsse:Security>
</soap:Header>
Is there any setting has to be done?
Thanks,
AP

Move Rules from DV to PR in CC5.2

We are in the process of converting from CC4.0 to CC5.2. In 4.0 we currently create our Rules using CC Rule Architect in DV, then transport to PR.
Once we get CC5.2 up and running, we will still create our Rules in DV. Can someone explain the process for getting the Rules from DV to PR?

I think the best approach either doing a full initial load or during the change management process is the solution brought by Naveen, the export/import rule functionnality in CC 5.2 should solve this issue.
I'm still working on the business processes that will support my current installation and havent tried the fct yet, but if this function allows to overwrite the old rule set with a fresh one after the import, it should the trick perfectly since this function was put in place to replace the ABAP transports.
Simpliest process should be to maintain in DEV and do first run of testing, Export to text file, import in QA, final test and signoff then export and final import in PRD.
Cheers
Jay

Get string inside brackets - Using Scan from string

I have a very basic question here. I would like extract a portion of a string which is inside brackets (Example: I work with [National Instruments]). In the string I only want to extract National Instruments, I do know logic to extract by finding the brackets and extracting them. But I am interested in using the Scan from string or possibly any other primitive in one shot. Please suggest if anyone has any idea, I am really not use to the format specifiers. I got this link for reference but I am not able to figure out a solution.
The best solution is the one you find it by yourself
Solved!
Go to Solution.

Hi,
I cannot say that it is the best solution, I use two Scan From String functions instead of one but it does work, so far ;-)
I use Characters in set.
I could not attach VI so FP is below.
Duri

Exclude Matches between records from a Particular Source system EDQ+Siebel UCM

I have a requirement to modify the EDQ-CDS Individual Match processor.
I want to exclude the match between two records (Driver and Candidate) if they have same value in a particular field and i want the Value to be hard-coded or specified as reference Data in EDQ (eg: if both driver and candidate are from same source say 'Source ABC' only then NO MATCH , for all other source records match should happen).
We tried adding Match rule which does the 'Exact Match' comparison on the field 'source' and set the rule to 'NO MATCH' , but this is excluding the matching between records from any system, which is not my requirement. I am only looking to exclude the match when the Source value is say 'Source ABC'.
I think it should be possible with the Comparison type 'In Value' or 'In List ' or 'All in List'. But not sure who to use these comparison rules and there is not much documentation available EDQ online.
Any help on this appreciated.
We are on Siebel 8.2.2.4 + EDQ 11.1.1.7 on Solaris 64 bit.

In List is documented in the 12.1.3 online help. Before this, it was an extension to the product used by CDS.
It is very straight-forward. You add it as a comparison, and create a reference data set with your list of values in it. If the identifier value hits the list, the comparison evaluates to true.
I've pasted most of the help from 12.1.3 below:
Comparison: In List
The In List comparison provides a way of making the application of a Match rule conditional on one or both identifier values in a comparison being matched with a single value or list of values.
Use
Use this comparison as a way to apply a Match rule only to a subset of the data in the Match process. For example:
Where a Match rule is designed to match Middle Eastern names, only apply it if a Country identifier matches a list of Middle Eastern countries.
Where a Match rule is designed to match Electronic products, only apply it if a Product Category identifier matches a single value.
The comparison may also be used to eliminate matches that would otherwise be caught by Match rules lower in the decision table, if used in a Match rule with a 'No Match' result; for example, so that matches on certain rules are only presented if Country or Nationality values are not in a list of safe values.
Options
Option
Type
Purpose
Default Value
Require data in both records?
Yes/No
If Yes, both input identifier values must contain data. If they do not, the comparison will always be 'False'.
If No, only one input identifier value must contain data.
No
Match whole value?
Yes/No
If Yes, the whole identifier value (or values) specified must match.
If No, tokens within the identifier will be matched against the list. In this case, delimiters must be specified in the relevant fields below to determine how to split tokens.
No
Required value reference data
Reference Data
A Reference Data set with a list of values, for example a set of country codes.
Clear
Required value
Free text
A single value to match against. Note that if a Reference Data set and a value is specified, both are matched against.
Clear
Require match for all values?
Yes/No
If Yes, all tokens in the identifier value(s) must match against the required list or value.
If No, any one of the tokens must match.
This option is only used if Match whole value? is set to "No".
No
Delimiter characters reference data
Free text or browse
A Reference Data set with a list of delimiter characters used to tokenize identifier values before matching against the list.
This option is only used if Match whole value? is set to "No".
None
Delimiter characters
Free text
This field is used to specify the delimiter characters to use as an alternative to linking to a Reference Data set.
This option is only used if Match whole value? is set to "No".
Note: If a Reference Data list of delimiters and specific characters are entered here, both are considered delimiter characters.
None

Generate best matching regex/rule from string set

Similar Messages

Maybe you are looking for