I've searched the forum and only found one thread which covers this but doesn't help me.
I have a table with peoples individual information in it.
Every person has a unique ID number.
I want to find the missing ID numbers when I delete a person so I can keep the numbers in sequence.
I am not using autonumber. The reason for this is coloured wristbands. Each wristband has a unique number and they are broken into 4 colours. each colour has a number range 1 - 300, 301 - 600, 601 - 900, 901 - 1200. Each person in the table has to have an ID number matching a wristband. The people are also broken into 4 colour categories. This means that not every range is filled before i need to start using the next range.
Hello, I have no idea what to search for to see if someone else has already asked this question. I've tried several things with no luck. So here goes...
I work in a library. We deal with thousands of journal subscriptions--we cancel our subscriptions and order new journals all of the time. Every paper journal we have is assigned a unique number, called a ZP number. When we throw out a journal, that ZP number can be assigned to a new journal.
I would like to find the earliest missing number in the sequence without having to check manually. Is there a way I can do a query or something to find missing numbers in the sequence? The ZP numbers start with 1 and go up to 9999.
So, I have two fields: Title and ZP#.
How can I do this?
This is a cloned table, used solely for assigning ZP#'s, so it can be altered in any way--including adding records with blank titles for each missing number.
I have a MS Access 2000 database with 136 data tables in it. What I would like to do is execute a piece of VBA code which will list for me in another table, (for example: Field1: TABLE NAME Field2: FIELD NAME 1 Field2: FIELD NAME2 etc), starting left to right, how many fields would have to be combined in each table to represent a unique record.
For instance:
TYPE ID TEXT 1 1 "Cats" 1 2 "Dogs" 2 1 "Rabbits"
In this example a combination of the fields TYPE and ID give a unique record.
I work for a non profit agency, and we offer a Representative Payee service (paying bills for those who can't handle their own money). I've created a database that keeps track of everyone's finances, and gives us all the reports we need. The only thing I have not been able to do is to actually print the checks.
I have a report that pulls up the data (in check format ... including making $1.00 say one dollar and 00/100). But I'm stuck trying to give the checks a check number. I can't figure out how to pick a number to start with, and then add numbers sequentially.
First I want to thank everyone who sent me a response to my original problem. Unfortunately I still have not been able to get the results that I need. So for those new individuals who may have missed my original thread I will try again:
I have 1 table (Student Records) that looks like this:
Last Name, First Name, Class Title, Completion Date
And I have a 2nd Table (Course List) that looks like this:
Class ID, Class Title, Frequency
What I am looking for is a query to tell me which of the classes in the table Course List are not in Student Records Table. Basically I needto know which students haven't completed a course.
I have a table of order lines. Each line of data the order number is repeated for every line. I need to just pull out one line for each order number. I assume a query would do this but I don't know how to create it. The order number is numeric. Can some one show me the way?
I have a sequencial number in my data base. It counts 1,2,3 all the way to 302344. I want to find missing records. How would I come up with the missing records in a separte database. In other words if I had 1,2,4,5,8,9. I would want to have a database that would show me 3,6,7 since they are missing.
I need to create a very simple database that would just store records and produce a couple of reports. I have three tables: one with the roster , one has records of the inventory items people from the roster receive and another one contains types of inventory we have. Everything is very simple except for one part. Every time we make a record of an inventory item given to someone, it requires not only employee id and inventory type from the existing tables. It needs us to enter a serial number of an item. This serial number contains a letter and a number.
Looks something like this - M100. Many people receive a consecutive set of inventory items. For example, from M100 to M150. There is no way to have a separate table with all serial numbers because they constantly change. That is why we need to have two text boxes that would allow us to input a range of serial numbers or just one number. Then the program should separate numbers from letters, evaluate the range, create new records of numbers and then put new numbers and a letter back together into one field in the table where we have all inventory records.
I found the following code online that allows me to find numbers within a range, but it only works for numbers.
Dim varRange As Variant Dim lngLow As Long Dim lngHigh As Long Dim lngCounter As Long DoCmd.Hourglass True
[code]...
I then found a piece of code that is supposed to separate numbers from letters, but I can't find a way to make it work.
Public Function FindNum(strName As String) As String Dim strTemp As String Dim i As Integer For i = 1 To Len(strName) strTemp = Mid(strName, i, 1) If (Asc(strTemp) < 91 And Asc(strTemp) > 64) Or (Asc(strTemp) < 128 And Asc(strTemp) > 96) Then FindNum = Right$(strName, Len(strName) - i) End If Next i End Function
how to alter the code to make it work for my specific situation.
I have a database with around 5083 records and I am trying to find the three most common numbers called in any one record such as 5 22 and 39 has been called together 50 times
Here is the structure # 1 through 5 is a number between 1 and 39 but never duplicated in same record
Record 1 |draw#|Date|#1|#2|#3|#4|#5| Record 2 |draw#|Date|#1|#2|#3|#4|#5| Record 3 |draw#|Date|#1|#2|#3|#4|#5| Record 4 |draw#|Date|#1|#2|#3|#4|#5| Record 5 |draw#|Date|#1|#2|#3|#4|#5| Record 6 |draw#|Date|#1|#2|#3|#4|#5| Record 7 |draw#|Date|#1|#2|#3|#4|#5| and so on for 5083 records
I have a list of what should be sequentaia numbers, but I am finding instances were a number seems to be missing..Is there a way to query the whole dataset and find all missing instances without having to trawl he list?
I am trying to create a query to find missing sequential numbers in a text field. I am using this specific field as a case number which is designated as two letters, the # sign, two digits indicating the year, a dash, and then a four digit number; For Example: AB#13-1234.
The reason for this query is to tell the user of this database that a specific case number has yet to be entered and needs to be. The case numbers are unique and will never be referenced more than once.
My table name is "MainDataTbl" and the field i'm trying to find the missing case numbers is titled "CaseNumber".
This is an ongoing problem I have had for 4 weeks now.
I have made a a system thats acts like a clock In/clock out Out system.
the structure is somthing like this ID Username tblDailyLog TimeIn MorningBreakOut MorningBreakOut LunchOut LunchIn AfternoonOut AfternoonIn TimeOut
All fields apart from ID (autonumber) and username (String*255) are Date field (there are a few others like DateOfTimesheet etc but they arnt important here)
When a user arrives in the morning they make a record which they use for the day
They then have a form with a whole bunch of buttons which simply updates the correct field. For example they click the "Sign in for the Day" button and it updates the correct field with the current time.
Everything was going fine until people noticed that every now and again a sign in time dissapeared.
I have hacked myself to death trying to solve this problem but still the updates go Astray.
Now each time a time is updated the process goes somthing like this
1. the user opens their timesheet for the day (the RS is SNAPSHOT and no locks)
2. User Hits a sign in/out button 3. The record source is changed to "" and all buttons hidden (to ensure the record isnt locked and to make sure you dont do two things at once) 3. The table is updated with the new time (using some dynamic SQL) 4. The table is repeatadly checked using a DO loop to make sure the the correct time went in. 5. when the returned time value of the field matches the varaible used to update it, the form is returned to normal and the user carries on his/her merry way (if it never matches the screen should crash but this never happens). 6. A New record is added to another table called "tblbugfixinglog" which records which field was updated and when. This is so that I have two records in two different ways (figured if one went astray I could pull it back off the other) 7. Another new record is added to yet another table called tblSQLRecord, which simply logs all .RUNSQL statements that are executed.
I thought that the two extra tables (and the check that the record had been updated) would help me track down where the records are going missing, but this isnt the case.
Now it appears that some records arnt being added to tblBugFixingLog and to tblSQLRecord either and some of these tables are getting quite a few #ERROR's in them..
None of the tables are related to any other and i've no idea how #ERROR lines are appearing in a table that has 1 function... to recieve new records ... no editing, no viewing, no deleting.
Does anyone have any idea how these updates/inserts can go missing or create #ERRORs. I've built plenty of Databases in my time and have never come across this. __________________________________________________ ______________
This is the function I use to add a record to tblBugfixingLog and tblSQLRecord
Private Sub AddBugLog(ByVal TimesheetNumber As Long, ByVal FieldUpdating As String, ByVal NewFieldValue) Dim TempSQL As String TempSQL = "INSERT INTO tblBugFixingLog (TimeAndDateOfEntrySERVER,TimeAndDateOfEntryPC,Fie ldUpdated,NewEntry,UserID,TimesheetNumber,Computer AssetNo) VALUES (" & _ "#" & Format(ServerGetTime(Environ$("LOGONSERVER"))) & "#," & _ "#" & Now & "#," & _ "'" & FieldUpdating & "'," & _ "'" & NewFieldValue & "'," & _ "'" & GetNTUser & "'," & _ "'" & TimesheetNumber & "'," & _ "'" & fOSMachineName & "')" ' MsgBox TempSQL DoCmd.RunSQL "INSERT INTO tblSQLRecord (Username,DateAndTime,Screen,TheSQL) VALUES('" & LoginInfo.sUsername & "','" & CStr(Now) & "','Add Bug Log function','" & CleanData(TempSQL) & "')", False 'CleanData is a function that removes ' and " from the SQL string so i can easily add the SQL string into the table DoCmd.RunSQL TempSQL, False End Sub
Public Function CleanData(ByVal DataToClean As String) Dim TempData As String Dim i As Integer TempData = "" For i = 1 To Len(DataToClean) Select Case Mid(DataToClean, i, 1) Case "'" TempData = TempData & "`" Case """" TempData = TempData & "`" Case Else TempData = TempData & Mid(DataToClean, i, 1) End Select Next i CleanData = TempData End Function
I'm having multiple problems with my database like things such as -
i'm currently working on the Query 2 - On the Phone database (ignore Query 1) and i want to search for multiple plot numbers preferably in one parameter prompt with a comma to seperate numbers. (this could be a multitude of numbers so i would like to be able to input as many as needed). Also when i do search on this query since the Criteria is a 'Between' Value i would expect everything between the 2 numbers input to show up - but a lot of numbers out of the range show up too - why is this? (The Numbers are like "69 to 136" and they will show up - but 1-69 and 136-170 would too
I would also like to implement the search results from Query 2 into the Form i currently have made but it just opens up a access table when the search is made?
i cannot link my database as it is too big for the server - But here are the Criteria for Query 2:
Plot No - (criteria = Between [Enter First Plot No:] And [Enter Last Plot No:]) Site - (criteria = Like "*" & [Enter Site:] & "*") Product - (criteria = Like "*" & [Enter Product:] & "*"
The Query is the one im most concerned about , i can live without a form.
Ok so in excel I have some numbers that are stored as text. The reason being that they are zip codes and some begin with 0 and excel doesn't want numbers to start with 0....so when I import these into an access field that has an input mask for zip codes...will it convert these correctly since the field is a text with input mask?
I have 2 fields that I would like to automate if possible
One field is called "p/o number" and another field called "line no"
These fields are part of an ordering database
Let say I have 200 items to purchase form 10 suppliers
And form example 20 items from each supplier
What I do at present is put the order number on each line item and the line number
example
p/o number line no
1 1 1 2 1 3
2 1 2 2 2 3 2 4
What I want to do is just put the first po number in the required line . Put the first line number in i.e. "1" and the macro will complete all the p/o numbers and line numbers for me as per the ones marked in red.
I have duplicate loan numbers in a table. I want to create a query that will assign a number so the the duplicates will have a new field called sequence. example
I have encountered some issues with the dates. For e.g, Jan-06, Feb-06, April-06, May-06 Is there any way to determine that Mar-06 is missing from the list. The listing will be growing.
Cause I need to have a culmulative counter so Jan-06 to Feb-06 will be counted as 2. But if there is a break in between, in this case, Mar-06, the counter will be set to 0, so for Apr-06, the counter will be 1.
Have been troubled by this issue for a long time, I appreciate all help given, thanks!
Hi... i have a excel file..... from which i have imported records.... after importing.... the records sequence in my table has changed..... can anyone tell me how to preserve the sequence that i had in excel..........
Not sure whether I am in the right forum. It is Table, Form and Query.
Have a table which index to sort on 4 fields - Proj, SubProj, Team, SubTeam.
1. I used this Table on a form, the data seems to be in random sequence instead of above.
2. Then I tried to sort it in a query which showed up fine ( as required) in the actual running of the query. However when I used this query in a form, the data doesn't sort as expected.
3. With above 1 and 2, I have also tried the Order Of in the Form properties which still do not do the trick.
can anyone shed somelight on this please.
BTW, After I first created the form, I just rename the record source in form property. Don't know whether this make any differences in the above result.
I tried this question a few weeks ago, but I'm trying again.
I have 2 Access 2002 files. One has a sort order of ascii and one has a sort order of international. I need them to both have the same sort order. It doesn't matter which one. Right now I get the above error message on one of them.
How can I change the sort order so that I can import Paradox files like I used to? :eek:
Hi, Its good to be back after a long time. Hope everyones fine. I have a query and would appreciate any help.
I have a master-table (tblM) and two related tables (tbl1, tbl2) with one-to-one relation.
Relation1 : tblM.ID (autonumber) related to tbl1.caseID (number;LongInteger) Relation2 : tblM.ID (autonumber) related to tbl2.caseID (number;LongInteger)
Relations are cascaded (referential integrity imposed).
I import data from excel and feed fields of tblM (I dont feed ID field from excel as it generates autonumber). At a time I feed 150 records (daily).
I have a front-end based on query from three underlying tables. Data-entry-operators do not work on tblM data as it is readonly in the front-end. They enter data in the tbl1 and tbl2 columns.
I noticed as they start entering data in tbl1 columns, the tbl1.caseID column gets data from tblM.ID on its own. Same happens in the case of tbl2.
My observation is : I find that tblM.ID is not equal to the number of records available in tblM. (ie tblM.ID is more/less than number of records in the table. So the tblM.ID generated is not in serial.
Next time as I start importing data from excel file to tblM it gives me error and not allowing further import.
I initially thought I could make an IIF expression that basically says: if [d] in this row is the same as [d] in the previous row then add 1 to the previous record [new number], otherwise, enter 1 in this record [new number] (thereby resetting it)....
I have searched the forums for 'numbered sequences' (amongst other things), and it appears that either the Dcount or Dmax function appear to be the solution, but I'm having some trouble applying it to this problem. I was hoping someone can help me through it.