Hulp nodig voor de toevoeging van drie nieuwe kolommen in bestaande bestand van fatching gegevens uit bestand - Page 3 - De Unix-en Linux Forum

Sandeep_Malik · #15 (**permalink**) 08-28-2008

Geef lijn per lijn op begrip van dit script, dat echt helpt me.

aigles · #16 (**permalink**) 08-28-2008

Code:

#
# DCDB - This block select all DCBEntry records
# and memorize Folder and TZ.
# => Folder[dcdb] and Tz[dcdb]
#

/^<DCDBEntry / {                    # Select DCDBEntry records
   split($0, dcdb, /"/);            #    Split record into dcdb array (sep=")
                                    #      dcdb[2]=dcdb, dcdb[4]=folder, 
                                    #      dcdb[6]=tz
   sub(/ *$/, "", dcdb[2]);         #    Remove trailing spaces from dcdb
   Folder[dcdb[2]] = dcdb[4];       #    Memorize folder for dcdb
       Tz[dcdb[2]] = dcdb[6];       #    Memorize tz for dcdb
   next;                            #    Proceed next input record
}                                   #

#
# LDS - This block select all LDSEntry records
# and memorize SiteUnit and Devide for folder,tz
# => SiteUnit[folder,tz] and Devicefolder,tz
#

/^<LDSEntry / {                     # Select LDSEntry records
   split($0, lds, /[":]/);          #    Split record into lds array (sep=" or sep=:)
                                    #      lds[2]=folder, lds[3]=siteunit
                                    #      lds[4]=device, lds[6]=tz
   folder = lds[2];                 #    Get folder
   tz     = lds[6];                 #    Get tz
   SiteUnit[folder, tz] = lds[3];   #    Memorize siteunit for folder,tz
     Device[folder, tz] = lds[4];   #    Memorize device for folder,tz
   next;                            #    Proceed next input record
}                                   #

#
# Input - These blocks proceed inputfiles
#

/^</ {                              # Select (and ignore) records from DCDB and LDS 
                                    # (DCDBEntry ans LDSEntry have already been processed)
   next;                            #    Procced next input record
}                                   #

FNR==1 {                            # Select first record of inputfile (logically all other files
                                    # have been processed in previus blocks).
   print $0, "Folder", "SU", "Dev"; #    Print first record (headers) and add new headers
   next;                            #    Proceed next input record
}                                   #

{                                   # Select all reamaining inputfile records
   dcdb_in = $2;                    #    Get dcdb from input record
   folder = Folder[dcdb_in];        #    Get memorized folder for this dcdb
   tz     =     Tz[dcdb_in];        #    Get memorized tz  for this dcdb
   su     = SiteUnit[folder, tz];   #    Get memorized siteunit for this forder,tz
   dev    =   Device[folder, tz];   #    Get memorized device for this forder,tz
   print $0, (folder ? folder : "?"), (su ? su : "?"), (dev ? dev : "?"); # Print record,
                                    #    folder, siteunit and device (unknown values are
                                    #    replaced by ?
   next;                            #    Proceed next input record
}                                   #

Sandeep_Malik · #17 (**permalink**) 08-29-2008

Thanks a lot Jean

Sandeep_Malik · #18 (**permalink**) 09-02-2008

Niet verplicht deze tijd

aigles · #19 (**permalink**) 09-02-2008

Weet u zeker dat alle gegevens die beginnen met '<' in de dossiers DCDB.xml en LDS.xml?
Ik heb het getest becommentarieerde awk script zonder probleem:

Code:

> cat DCDB.xml
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<DCDBTable>
<DCDBEntry DCDB="0862976 " folder="768678" timeZone="2"/>
<DCDBEntry DCDB="0911297 " folder="975426" timeZone="1"/>
<DCDBEntry DCDB="0201347 " folder="1389781" timeZone="2"/>
<DCDBEntry DCDB="0800659 " folder="2035595" timeZone="4"/>
<DCDBEntry DCDB="0123033 " folder="2143699" timeZone="2"/>
<DCDBEntry DCDB="0911515 " folder="2315643" timeZone="1"/>
<DCDBEntry DCDB="0123913 " folder="2367867" timeZone="2"/>
<DCDBEntry DCDB="0713934 " folder="2407712" timeZone="5"/>
</DCDBTable>
> cat LDS-1.xml
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<LDSMappingTable>
<LDSEntry LDSKey="3351216:1:TV" LDSValue="7"/>
<LDSEntry LDSKey="1389781:1:VCR1" LDSValue="2"/>
<LDSEntry LDSKey="3351216:2:TV" LDSValue="2"/>
<LDSEntry LDSKey="3351216:3:TV" LDSValue="6"/>
<LDSEntry LDSKey="3351511:1:TV" LDSValue="10"/>
<LDSEntry LDSKey="3351511:1:VCR1" LDSValue="11"/>
<LDSEntry LDSKey="3351511:2:TV" LDSValue="4"/>
<LDSEntry LDSKey="3351511:2:VCR1" LDSValue="5"/>
</LDSMappingTable>
> cat LDS-2.xml
<?xml version="1.0" encoding="UTF-8" standalone="yes"?>
<LDSMappingTable>
<LDSEntry LDSKey="3351217:1:XTV" LDSValue="7"/>
<LDSEntry LDSKey="1389782:1:XVCR1" LDSValue="2"/>
<LDSEntry LDSKey="3351217:2:XTV" LDSValue="2"/>
<LDSEntry LDSKey="3351217:3:XTV" LDSValue="6"/>
<LDSEntry LDSKey="3351512:1:XTV" LDSValue="10"/>
<LDSEntry LDSKey="3351512:1:XVCR1" LDSValue="11"/>
<LDSEntry LDSKey="3351512:2:XTV" LDSValue="4"/>
<LDSEntry LDSKey="3351512:2:XVCR1" LDSValue="5"/>
</LDSMappingTable>
> cat inputfile
1DATE HHIDLDS BothOff 91 NG A=D 90faulRoundOnOff OffOn OthersMATCHED NOMATCH MATCH%
0721 0201136 1 544 . . . 1 . . . 895 1 99.89
0721 0201347 1 1296 . . . . . . . 144 . 100.0
0721 0201347 2 818 . . . . . . . 622 . 100.0
0721 0201364 1 1123 . . . . . . . 317 . 100.0
0721 0201364 2 1327 . . . . . . . 113 . 100.0
> cat Malik.awk
#
# DCDB - This block select all DCBEntry records
# and memorize Folder and TZ.
# => Folder[dcdb] and Tz[dcdb]
#

/^<DCDBEntry / {                    # Select DCDBEntry records
   split($0, dcdb, /"/);            #    Split record into dcdb array (sep=")
                                    #      dcdb[2]=dcdb, dcdb[4]=folder,
                                    #      dcdb[6]=tz
   sub(/ *$/, "", dcdb[2]);         #    Remove trailing spaces from dcdb
   Folder[dcdb[2]] = dcdb[4];       #    Memorize folder for dcdb
       Tz[dcdb[2]] = dcdb[6];       #    Memorize tz for dcdb
   next;                            #    Proceed next input record
}                                   #

#
# LDS - This block select all LDSEntry records
# and memorize SiteUnit and Devide for folder,tz
# => SiteUnit[folder,tz] and Devicefolder,tz
#

/^<LDSEntry / {                     # Select LDSEntry records
   split($0, lds, /[":]/);          #    Split record into lds array (sep=" or sep=:)
                                    #      lds[2]=folder, lds[3]=siteunit
                                    #      lds[4]=device, lds[6]=tz
   folder = lds[2];                 #    Get folder
   tz     = lds[6];                 #    Get tz
   SiteUnit[folder, tz] = lds[3];   #    Memorize siteunit for folder,tz
     Device[folder, tz] = lds[4];   #    Memorize device for folder,tz
   next;                            #    Proceed next input record
}                                   #

#
# Input - These blocks proceed inputfiles
#

/^</ {                              # Select (and ignore) records from DCDB and LDS
                                    # (DCDBEntry ans LDSEntry have already been processed)
   next;                            #    Procced next input record
}                                   #

FNR==1 {                            # Select first record of inputfile (logically all other files
                                    # have been processed in previus blocks).
   print $0, "Folder", "SU", "Dev"; #    Print first record (headers) and add new headers
   next;                            #    Proceed next input record
}                                   #

{                                   # Select all reamaining inputfile records
   dcdb_in = $2;                    #    Get dcdb from input record
   folder = Folder[dcdb_in];        #    Get memorized folder for this dcdb
   tz     =     Tz[dcdb_in];        #    Get memorized tz  for this dcdb
   su     = SiteUnit[folder, tz];   #    Get memorized siteunit for this forder,tz
   dev    =   Device[folder, tz];   #    Get memorized device for this forder,tz
   print $0, (folder ? folder : "?"), (su ? su : "?"), (dev ? dev : "?"); # Print record,
                                    #    folder, siteunit and device (unknown values are
                                    #    replaced by ?
   next;                            #    Proceed next input record
}
> awk -f Malik.awk DCDB.xml LDS-1.xml LDS-2.xml inputfile
1DATE HHIDLDS BothOff 91 NG A=D 90faulRoundOnOff OffOn OthersMATCHED NOMATCH MATCH% Folder SU Dev
0721 0201136 1 544 . . . 1 . . . 895 1 99.89 ? ? ?
0721 0201347 1 1296 . . . . . . . 144 . 100.0 1389781 1 VCR1
0721 0201347 2 818 . . . . . . . 622 . 100.0 1389781 1 VCR1
0721 0201364 1 1123 . . . . . . . 317 . 100.0 ? ? ?
0721 0201364 2 1327 . . . . . . . 113 . 100.0 ? ? ?
>

Sandeep_Malik · #20 (**permalink**) 09-09-2008

Niet verplicht deze tijd

aigles · #21 (**permalink**) 09-09-2008

Code:

nawk -f Test.awk DCDB.xml LDS.xml input > input_new.txt

Thread Tools	Zoeken in deze Thread
Toon Printversie E-mail deze pagina	Zoeken in deze Thread: Uitgebreid zoeken
Display Modes	Beoordeel deze draad
Lineaire modus Switch to Hybrid Mode Switch to Threaded Mode	Beoordeel deze draad:

Meer UNIX en Linux Forum Onderwerpen Misschien vindt u Helpful
Draad	Thread Starter	Forum	Antwoorden	Last Post
Greping kolommen gegevens uit het bestand.	esungoe	Programmeren en Shell Scripting	6	08-05-2008 09:11
Het toevoegen van kop naar een bestaand bestand	shash	UNIX voor Dummies Questions & Answers	5	07-24-2008 08:46
Het toevoegen van kolommen in een bestand	Figaro	UNIX voor Dummies Questions & Answers	5	07-21-2008 02:50
Perl: het toevoegen van kolommen in het CSV-bestand met informatie in elke	dolo21taf	Programmeren en Shell Scripting	1	03-05-2008 02:52
Noodzaak voor het toevoegen van een lijn van de gegevens van reeds bestaande bestand in Unix ..	charan81	Programmeren en Shell Scripting	4	01-21-2006 03:31