search txt file for a certain number

[durgensrun]

I am trying to search a text file for data. Each data line has 10 items. the first data item is a 6 digit id number.
I input an id number to search for, then print all 10 items on the screen on that line. I think I am missing something.
thanks

Code: QB64: [Select]

1. TYPE ContactType 'there are 10 items in the text file
2.     idnum AS STRING * 6
3.     consumerid AS STRING * 6
4.     firstname AS STRING * 20
5.     lastname AS STRING * 20
6.     address1 AS STRING * 25
7.     address2 AS STRING * 25
8.     city AS STRING * 20
9.     state AS STRING * 2
10.     zip AS STRING * 5
11.     phone AS STRING * 10
12.     email AS STRING * 25
13. END TYPE
14. DIM consumer AS ContactType
15.  
16. CLS
17. LOCATE 5, 30: INPUT "consumerid: ", idnum
18. OPEN "durgenDB/contactlist.txt" FOR BINARY AS #1
19. a$ = SPACE$(LOF(1))
20. GET #1, 1, a$
21. CLOSE #1
22.  
23. foundError = INSTR(a$, "idnum")
24. IF foundidnum THEN
25.  
26.     DO
27.         PRINT consumer.consumerid, consumer.firstname, consumer.lastname, consumer.address1, consumer.address2, consumer.city, consumer.state, consumer.zip, consumer.phone, consumer.email
28.         foundidnum = INSTR(foundidnum + 1, a$, "idnum")
29.     LOOP WHILE foundidnum > 0
30. END IF

[SpriggsySpriggs]

If I were you I'd do a DO, LINE INPUT, LOOP rather than getting it all at once. That way you can stop once you find the value you are looking for.

[durgensrun]

this is what's in the text file, the data in it is only temporary for testing.

"123456","ZEF                 ","DURGEN              ","MOUNT                ","                         ","REYNO               ","AR","12345","0123456789","moo@zap.com              "
"098765","BOB                 ","MOLE                ","KIM DR                   ","                         ","BIGGERS             ","AR","65432","1234567890","bee@mook.com             "

the first numbers, a 6 digit number, is the unique identifier for that line of data.
what I need to do is search the text file for that unique number, and print all data items on the screen on that line only.
thanks

[bplus]

OMG you set up a record to a data file!

You can make an array of these records, your user defined type, and access each field of each index item of an array of records.

Code: QB64: [Select]

1. TYPE ContactType 'there are 10 items in the text file <<< VVV these are fields of a record with Random access
2.     idnum AS STRING * 6
3.     consumerid AS STRING * 6
4.     firstname AS STRING * 20
5.     lastname AS STRING * 20
6.     address1 AS STRING * 25
7.     address2 AS STRING * 25
8.     city AS STRING * 20
9.     state AS STRING * 2
10.     zip AS STRING * 5
11.     phone AS STRING * 10
12.     email AS STRING * 25
13. END TYPE
14. DIM consumer AS ContactType  ' a record
15.  
16. CLS
17. LOCATE 5, 30: INPUT "consumerid: ", idnum
18. OPEN "durgenDB/contactlist.txt" FOR RANDOM AS #1 LEN = LEN(consumer)
19. WHILE NOT EOF(1)
20.     recNumber = recNumber + 1
21.     GET #1, recNumber, consumer
22.     IF VAL(consumer.idnum) = idnum THEN ' found!
23.         'print consumer.this and that as needed
24.  
25.          EXIT WHILE
26.     END IF
27. WEND
28.  

I am rusty with Random access but this might work.

EDITS: sorry still fixing typos

You can also input the idnum as a string instead of a number might be better for matching.

[durgensrun]

bplus, I am getting error 52 on line 19, "WHILE NOT EOF(1)"
thanks

[bplus]

Hi @durgensrun

Did you make all your strings as long as spec'd in the Type? If any one is one byte short you won't load a single record.

Yeah also we should count number of records in file first by taking LOF(1) and dividing by record len and exit if we try to go get a record past the last record number.


Quote
"123456","ZEF                 ","DURGEN              ","MOUNT                ","                         ","REYNO               ","AR","12345","0123456789","moo@zap.com              "
"098765","BOB                 ","MOLE                ","KIM DR                   ","                         ","BIGGERS             ","AR","65432","1234567890","bee@mook.com             "

Yeah a Random Access file is not full of commas and double quotes between fields.

[durgensrun]

Hi bplus
It's been 30 years since I made my first program, I am relearning with a special project. I am making a small relational 3-page database.
the text file I posted will hold all the names and addresses and would be added to, so I don't know how many lines will be on it, same with the other 2 text files.

[bplus]

Do you know or is it remember about filling fixed string fields for a record?

Wait have you worked with Random Access?

Your file looks like CVS style data lines.

[durgensrun]

Long ago I did.

[bplus]

Yeah what you have is more CVS style except you don't need spaces for that.

A random access file would not have quotes or commas but allot of spaces! specially with fixed strings.

Random access you'd want to make numbers as much as possible at the smallest Type size that would fit your data range.

[bplus]

@durgensrun

Did you see this I made for badger for data entry for multiple fields to create / edit a record?

Code: QB64: [Select]

1. TYPE customer
2.     id AS LONG
3.     first AS STRING * 25
4.     last AS STRING * 25
5.     address AS STRING * 30
6.     city AS STRING * 25
7.     state AS STRING * 2
8.     zip AS STRING * 5
9.     phone AS STRING * 10
10. END TYPE
11.  
12. DIM SHARED nRecs ' for when you get a file going
13.  
14.  
15. DIM myRec AS customer 'edit this record over and over for testing
16. DO
17.     edit = EditCustomer(myRec, 1) ' assign to first record 1 if not Cancelled
18.     CLS
19.     IF edit THEN
20.         PRINT "Record #:"; myRec.id
21.         PRINT "   First:"; myRec.first
22.         PRINT "    Last:"; myRec.last
23.         PRINT " Address:"; myRec.address
24.         PRINT "    City:"; myRec.city
25.         PRINT "   State:"; myRec.state
26.         PRINT "   Phone:"; myRec.phone
27.     ELSE
28.         PRINT "user cancelled edit"
29.     END IF
30.     PRINT "(SLEEPing) press any to continue..."
31.     SLEEP
32.     CLS
33. LOOP
34. 'if edit then fileRec myRec, 1
35.  
36. 'return 0 if user cancelled, -1 if record edit was accepted
37. FUNCTION EditCustomer% (customerRec AS customer, recNumber AS LONG) ' use for new Customer or edit Customer
38.     DIM rec(1 TO 7, 0 TO 1) AS STRING
39.     rec(1, 0) = "First Name"
40.     rec(2, 0) = "Last Name"
41.     rec(3, 0) = "Address"
42.     rec(4, 0) = "City"
43.     rec(5, 0) = "State (abbrev:2)"
44.     rec(6, 0) = "Zip (5)"
45.     rec(7, 0) = "Phone" ' might split to Home and Mobile
46.     IF customerRec.first <> "" THEN rec(1, 1) = customerRec.first
47.     IF customerRec.last <> "" THEN rec(2, 1) = customerRec.last
48.     IF customerRec.address <> "" THEN rec(3, 1) = customerRec.address
49.     IF customerRec.city <> "" THEN rec(4, 1) = customerRec.city
50.     IF customerRec.state <> "" THEN rec(5, 1) = customerRec.state
51.     IF customerRec.zip <> "" THEN rec(6, 1) = customerRec.zip
52.     IF customerRec.phone <> "" THEN rec(7, 1) = customerRec.phone
53.     doAgain:
54.     CLS
55.     FOR i = 1 TO 7 'show what we have
56.         PRINT i, rec(i, 0); " = "; rec(i, 1)
57.     NEXT
58.     PRINT: PRINT "Press a to Accept, n for New, a digit to edit field, or escape to Cancel."
59.     choice$ = getChar$("aAnN1234567" + CHR$(27))
60.     IF ASC(choice$) = 27 THEN ' maybe this should be a function
61.         EditCustomer% = 0: EXIT FUNCTION 'cancelled  EDIT to 0
62.     ELSEIF UCASE$(choice$) = "A" THEN
63.         customerRec.id = recNumber 'assign the record
64.         customerRec.first = rec(1, 1)
65.         customerRec.last = rec(2, 1)
66.         customerRec.address = rec(3, 1)
67.         customerRec.city = rec(4, 1)
68.         customerRec.state = rec(5, 1)
69.         customerRec.zip = rec(6, 1)
70.         customerRec.phone = rec(7, 1)
71.         EditCustomer% = -1: EXIT FUNCTION 'done
72.     ELSEIF UCASE$(choice$) = "N" THEN
73.         CLS
74.         FOR i = 1 TO 7
75.             PRINT rec(i, 0); " > ";
76.             INPUT ""; rec(i, 1)
77.         NEXT
78.         GOTO doAgain
79.     ELSEIF INSTR("1234567", choice$) THEN
80.         CLS
81.         PRINT rec(VAL(choice$), 0); " > ";
82.         INPUT ""; rec(VAL(choice$), 1)
83.         GOTO doAgain
84.     END IF
85. END SUB
86.  
87. FUNCTION getChar$ (fromStr$) ' get a char$ usually for a menu
88.     DIM OK AS INTEGER, k$
89.     WHILE OK = 0
90.         k$ = INKEY$
91.         IF LEN(k$) THEN
92.             IF INSTR(fromStr$, k$) <> 0 THEN OK = -1
93.         END IF
94.         _LIMIT 200
95.     WEND
96.     _KEYCLEAR
97.     getChar$ = k$
98. END SUB
99.  
100.  

Modify it for your type and half the work of data entry is done! :)

[SMcNeill]

Question:  If this ID is unique, and doesn’t appear anywhere else, can’t you just:

OPEN file$ FOR BINARY AS #1
l = LOF(1)
temp$ = SPACE$(l)
GET #1, 1, temp$
CLOSE

p = INSTR(temp$, id$)
IF p <> 0 THEN ... you found your record at position p

EDIT: Since the ID is *always* in the starting position of your record, you can even error check to rule out false positives with a position check:
IF p MOD LEN(record_type) = 1 THEN .... it’s valid......  ELSE....  it’s in an address, or such and not in the proper position to be an ID field

[bplus]

If you were going Random Access route the customer ID would be same as record number, so you just GET record # of customer. Random Access depends upon records and fields of a record being exactly the same size so lots of empty spaces for all strings records/fields.

But I think you might like variable length fields and Steve had start of that somewhere, ah here:
https://www.qb64.org/forum/index.php?topic=3076.msg123525#msg123525
Steves method depends on keeping another file to record lengths of each and every field.

Another approach with variable length depends on using delimiters for records and another for fields in record. Use Split as parser, use of CHR$(10) would be good to separate records (called delimiter) and another character to delimit the fields (each item in your customer Type). You'd first SPLIT the file into records array and then each record split again by another delimiter (maybe comma or semi colon or colon?) and ideally your fields would be in same order.

Split1000 very powerful and handy parser: https://www.qb64.org/forum/index.php?topic=1607.0

And still your customer ID number could be same as record number = array index after split files into records for fast lookup of a customer.

sample file: fields comma delimited and records chr$(10) delimited:
fileStr$ = "1,ZEF,DURGEN,MOUNT, ,REYNO,AR,12345,0123456789,moo@zap.com" + chr$(10) +
"2,BOB,MOLE,KIM DR, ,BIGGERS,AR,65432,1234567890,bee@mook.com" + chr$(10) + ... 
customer ID = same as record #

in an editor chr$(10) is same as next line, so piece of cake to write the file in a standard text editor (no double quotes each line is record.)

Also with variable length strings you don't have to use TYPE with fixed strings but TYPE might be used with variable strings to keep better organized and labeled specially with 11 fields. Just have to JOIN fields to make a record and JOIN records to make a single giant string to store back as file. JOIN is opposite of split building strings with commas between each field the building file string with CHR$(10) between each record.

RANDOM ACCESS doesn't need the splits and joins nor another file to track lengths of ever individual field, it just uses allot empty space to ensure each field from Type is same size so records are same size so you can GET or PUT records a chunk at a time.

[durgensrun]

I been working on the problem. This is what I have so far. "sub finder" scans the file, when it finds a match it stops and says found it.
"Sub display", reads the 10 variables in the line, and prints them to the screen. The "IDnumber" is always the first variable in the set of 10.
The next problem is how to combine the 2 subs. The sequence I need is, I input an "IDnumber" to search for. Then scans the file. When that number is found.
The program goes and reads in the 10 variables including the "IDnumber" found on that line and prints them on screen.

I am thinking if I count the variables during the search this would give me an address of sorts, then rescan the file counting the variables and stoping the scan -1 prior to the found variable.
Then load in the next 10 variables I need to print.

Code: QB64: [Select]

1. SUB finder
2.     CLS
3.     OPEN "durgenDB/consumerprofile.dat" FOR INPUT AS #1
4.     INPUT " Profile number to find  ", searchnumber
5.  
6.     DO UNTIL EOF(1)
7.         INPUT #1, IDnumber
8.         PRINT IDnumber
9.         IF IDnumber = searchnumber THEN PRINT "found it   ": END
10.     LOOP
11.     CLOSE #1
12. END SUB
13.  
14. SUB display
15.     OPEN "durgenDB/consumerprofile.dat" FOR INPUT AS #1
16.     INPUT #1, IDnumber
17.     INPUT #1, FirstName$
18.     INPUT #1, LastName$
19.     INPUT #1, Address1$
20.     INPUT #1, Address2$
21.     INPUT #1, City$
22.     INPUT #1, State$
23.     INPUT #1, Zip$
24.     INPUT #1, Phone$
25.     INPUT #1, Email$
26.     CLOSE #1
27.  
28.     LOCATE 2, 10: PRINT "Costumer Profile": PRINT "-----------------------------------------------------------------"
29.     LOCATE 4, 10: PRINT "Consumer No.: ": LOCATE 4, 22: PRINT IDnumber
30.     LOCATE 6, 10: PRINT "Name: ": LOCATE 6, 16: PRINT RTRIM$(FirstName$); " "; RTRIM$(LastName$)
31.     LOCATE 8, 10: PRINT "Mailing Address: ": LOCATE 8, 27: PRINT RTRIM$(Address1$)
32.     LOCATE 10, 10: PRINT "Shipping Address: ": LOCATE 10, 28: PRINT "bloop" 'RTRIM$(Adress2$)
33.     LOCATE 12, 10: PRINT "City: ": LOCATE 12, 16: PRINT City$
34.     LOCATE 14, 10: PRINT "State: ": LOCATE 14, 17: PRINT State$
35.     LOCATE 16, 10: PRINT "Zip: ": LOCATE 16, 15: PRINT Zip$
36.     LOCATE 18, 10: PRINT "Phone: ": LOCATE 18, 17: PRINT Phone$
37.     LOCATE 20, 10: PRINT "E-Mail: ": LOCATE 20, 18: PRINT Email$
38. END SUB

[durgensrun]

Solved it, for the curious how I did it.

Code: QB64: [Select]

1.  
2. show
3.  
4. SUB show
5.     CLS
6.     OPEN "durgenDB/consumerprofile.dat" FOR INPUT AS #1
7.     INPUT " Profile number to find  ", searchnumber
8.     DO UNTIL EOF(1)
9.         INPUT #1, IDnumber
10.         IF IDnumber = searchnumber THEN
11.             INPUT #1, FirstName$
12.             INPUT #1, LastName$
13.             INPUT #1, Address1$
14.             INPUT #1, Address2$
15.             INPUT #1, City$
16.             INPUT #1, State$
17.             INPUT #1, Zip$
18.             INPUT #1, Phone$
19.             INPUT #1, Email$
20.         END IF
21.     LOOP
22.     CLOSE #1
23.     CLS
24.     LOCATE 2, 10: PRINT "Costumer Profile": PRINT "-----------------------------------------------------------------"
25.     LOCATE 4, 10: PRINT "Consumer No.: ": LOCATE 4, 22: PRINT searchnumber
26.     LOCATE 6, 10: PRINT "Name: ": LOCATE 6, 16: PRINT RTRIM$(FirstName$); " "; RTRIM$(LastName$)
27.     LOCATE 8, 10: PRINT "Mailing Address: ": LOCATE 8, 27: PRINT Address1$
28.     LOCATE 10, 10: PRINT "Shipping Address: ": LOCATE 10, 28: PRINT Address2$
29.     LOCATE 12, 10: PRINT "City: ": LOCATE 12, 16: PRINT City$
30.     LOCATE 14, 10: PRINT "State: ": LOCATE 14, 17: PRINT State$
31.     LOCATE 16, 10: PRINT "Zip: ": LOCATE 16, 15: PRINT Zip$
32.     LOCATE 18, 10: PRINT "Phone: ": LOCATE 18, 17: PRINT Phone$
33.     LOCATE 20, 10: PRINT "E-Mail: ": LOCATE 20, 18: PRINT Email$
34.  
35.     DIM hope AS STRING
36.     hope = UCASE$(hope)
37.     LOCATE 23, 10: INPUT "View another Y/N ", hope
38.     IF UCASE$(hope) = "Y" THEN
39.         show
40.     ELSE
41.         END 'menu
42.     END IF
43.  
44. END SUB
45.  
46.  
47.  
48.  
49.  
50.