PX file format 2013

Readers guide

What this instruction covers

This manual describes how a PC-Axis PX file is constructed. The manual is intended for people that wants to get a introduction of how to use and to extend the Core part of the PC-Axis SDK. The reader are not required have any type of PC-Axis experience to understand the content but it helps.

How this guide is organized

The document is divided into two part. The first part defines how the PX file structure works. The second part defined all keywords used in the PC-Axis File Format.

Related documents

Earlier versions of the PC-Axis file format.

Getting started

We must first try to explain the basic structure of the PX-file. PX-files consist of keywords that describes the contents or the data cube. Some keywords are mandatory while some are optional. Some keywords have an language option and some keywords have a sub key and there is always at least one value specified for a keyword. Some basic rules for how keywords are written - The keyword is mandatory. - Language is optional. - Language is always written within square brackets. - Language is always written immediately after the keyword. - The sub key is optional. - The sub key is always written within parentheses. - The sub key is written is written after language if present otherwise it is written after the keyword. - Value(s) are mandatory. - Value(s) are always written after the equal sign. - If there are multiple values each value is written within quotation marks and are separated by a comma sign. - If there is only one value then it can be written without any quotation marks. - Values(s) are delimited with quotation marks which means that a value itself cannot contain a quotation mark. - The end of values (and the beginning of a new keyword) is always marked with a semicolon. - Longer texts are divided into several records (except for value text orvariable names which must in one line when in parentheses after a keyword).

KEYWORD[LANGUAGE](SUB-KEY)="VALUE_1","VALUE_2", …;
KEYWORD[LANGUAGE](SUB-KEY)=VALUE_1;
KEYWORD[LANGUAGE]="VALUE_1","VALUE_2", …;
KEYWORD (SUB-KEY)="VALUE_1","VALUE_2", …;
KEYWORD[LANGUAGE]=VALUE_1;
KEYWORD(SUB-KEY)=VALUE_1;

The file itself is divided into two parts one metadata part and the data part. The data part starts after the keyword DATA. If the keyword CHARSET is missing it means that all texts are in DOS text format, so that the same files can be used both in the DOS and the Windows version of PC-Axis. Starting with version 2005 it is possible to have more than one language in a PX file. The second language is repeated for all necessary keywords. CONTENTS="Population";, CONTENTS[sv]=”Befolkning”; etc. Which languages are available are given in the keyword LANGUAGES.

Keywords

Alphabetical list of keywords

Keyword	Mandatory	Multiplicity	Language dependent	Default value
AGGREGALLOWED	No	0-1	No	YES
ATTRIBUTE-ID	No	0-1	No
ATTRIBUTE-TEXT	No	0-1	Yes
ATTRIBUTES	No	0-?	No
AUTOPEN	No	0-1	No	NO
AXIS-VERSION	No	0-1	No
BASEPERIOD	No	0-?	Yes
CELLNOTE	No	0-?	Yes
CELLNOTEX	No	0-?	Yes
CFPRICES	No	0-?	No
CHARSET	No	0-1	No
CODEPAGE	No	0-1	No	iso-8859-1
CODES	Yes	0-16	No
CONFIDENTIAL	No	0-1	No	0
CONTACT	No	0-?	Yes
CONTENTS	Yes	1	Yes
CONTVARIABLE	Yes	0-1	Yes
COPYRIGHT	No	0-1	No	NO
CREATION-DATE	Yes	0-1	No
DATA	Yes	1	No
DATABASE	No	0-1	Yes
DATANOTE	No	0-?	Yes
DATANOTECELL	No	0-1	Yes
DATANOTESUM	No	0-1	Yes
DATASYMBOL1	No	0-1	Yes	.
DATASYMBOL2	No	0-1	Yes	..
DATASYMBOL3	No	0-1	Yes	...
DATASYMBOL4	No	0-1	Yes	....
DATASYMBOL5	No	0-1	Yes	.....
DATASYMBOL6	No	0-1	Yes	......
DATASYMBOLNIL	No	0-1	Yes	-
DATASYMBOLSUM	No	0-1	Yes
DAYADJ	0-1	0-?	Yes	NO
DECIMALS	Yes	1	No
DEFAULT-GRAPH	No	0-1	No
DESCRIPTION	No	0-1	Yes
DESCRIPTIONDEFAULT	No	0-1	No
DIRECTORY-PATH	No	0-1	No
DOMAIN	No	0-16	No
DOUBLECOLUMN	No	0-16	Yes	NO
ELIMINATION	No	0-16	Yes	NO
FIRST-PUBLISHED	No	0-1	No
HEADING	Yes	(0)-1	Yes
HIERARCHIES	No	0-1	Yes
HIERARCHYLEVELS	No	0-1	Yes
HIERARCHYLEVELSOPEN	No	0-1	Yes
HIERARCHYNAMES	No	0-1	Yes
INFO	No	0-1	Yes
INFOFILE	No	0-1	Yes
KEYS	No	0-16	Yes
LANGUAGE	Yes	1	No
LANGUAGES	No	0-1	No
LAST-UPDATED	No	0-1	No
LINK	No	0-1	Yes
MAP	No	0-1	Yes
MATRIX	Yes	1	No
META-ID	No	0-?	Yes
NEXT-UPDATE	No	0-1	No
NOTE	No	0-?	Yes
NOTEX	No	0-?	Yes
OFFICIAL-STATISTICS	No	0-1	No	NO
PARTITIONED	No	0-16	Yes
PRECISION	No	0-?	Yes
PRESTEXT	No	0-16	No	1
PX-SERVER	No	0-1	No
REFPERIOD	No	0-1/0-?	Yes
ROUNDING	No	0-1	No
SEASADJ	No	0-?	No	NO
SHOWDECIMALS	No	0-1	No
SOURCE	No	0-1	Yes
STOCKFA	No	0-?	Yes
STUB	Yes	0-1	Yes
SUBJECT-AREA	Yes	1	Yes
SUBJECT-CODE	Yes	1	No
SURVEY	No	0-1	Yes
SYNONYMS	No	0-1	No
TABLEID	No	0-1	No
TIMEVAL	No	0-1	Yes
TITLE	Yes	1	Yes
UNITS	Yes	1-?	Yes
UPDATE-FREQUENCY	No	0-1	No
VALUENOTE	No	0-?	Yes
VALUENOTEX	No	0-?	Yes
VALUES	Yes	1-16	Yes
VARIABLE-TYPE	No	0-16	No

Mandatory keywords

These keywords are compulsory. The text (the contents belonging to the key word) should be written on the same line as the keyword or on a line of its own, always within quotation marks.

CONTENTS

Type: Text Length: 256 characters Multiline: No

Information about the contents, which makes up the first part of a title created when retrieving tables from PC-Axis. The text must not exceed 256 characters (before 2002 only 100 chars).

DATA

Type: integer/text Length: N/A Multiline: Yes

The keyword DATA must be placed at the end of the file, followed by all the data cells or if KEYS are used the variable values and all data cells that differ from 0.

File without KEYS

After the key word follows all data cells that belong to the table. In most cases the cells are written with one record per line in the table. The number of records will thus be determined by the number of values in the stub. The length of the records depends upon the number of values in the heading, and also upon the size of figures. Table cells that contain a dash, one, two, three, four, five or six dots should be within quotation marks. For data without keys it is possible to write all cells in just one record, terminated by a semicolon. PC-Axis accepts the delimiters comma, space, semicolon, tabulator. The different delimiters are synonyms and can be mixed in the file. Recommended delimiter is space.

Files with KEYS

For every variable in the stub is indicated the value for the variable within quotation marks, comma separated, followed by all data cells for that row (no quotation marks, space separated).

Whether the text from VALUES or from CODES are used for a variable is indicated by the keyword KEYS(”var”)=CODES or KEYS(”var”)=VALUES. It is possible to use VALUES from one variable and CODES for another in the same file. Rows that only contain 0 (zeros) are excluded from the file One - six dots Data that is missing is stored as one, two, three, four, five or six dots. It is not possible to get a dot or two dots by excluding a value. Thus, you cannot write 1,,2,3 in order to get 1,".",3,4 For PC-Axis version prior to 2005 a dash is presented as 0 and saved as 0. Thus, you cannot afterwards find out if the original value was - or 0. Starting with 2005 PC-Axis keeps track on which cells are – and the keyword DATASYMBOLNIL can be used to show something else in the table. How the dots are shown on screen is determined by the keywords DATASYMBOL1, etc. If these keywords are missing the presentation is taken from PC-Axis language files.

Number of decimals when a file is saved

The keyword DECIMALS determines the number of decimals to be saved. If the file also contains the key word PRECISION, PC-AXIS saves as many decimals as corresponds to the largest number specified by DECIMALS or PRECISION. All cells are stored with the same number of decimals.

DECIMALS

Type: integer [0, 15] Length: Lines of 256 characters Multiline: No

The number of decimals in the table cells. 0 - 15. (0-6 if SHOWDECIMALS is not included). Indicates how many decimals will be saved in the PC-Axis file. Written without quotation marks. Compare SHOWDECIMALS.

DESCRIPTION

Type: text Length: N/A Multiline: Yes

If a file contains DESCRIPTION, when fetching from a disk, this text is used to show the contents of the disk. C.f. TITLE. If the user wants to save a file in PC-AXIS and writes a text that describes the file, this text will be saved as DESCRIPTION. The text will be used to show the contents of tables in the SUBJECT AREA. The text is not presented as a note nor in any other way when the table is presented on the screen unless the keyword DESCRIPTIONDEFAULT is used. In this case the description is shown instead of the title.

HEADING

Type: text Length: 256 character per value Multiline: Yes

At least one of the keywords STUB or HEADING must be included. Usually both are included, as you choose one or several variables for the stub and the heading, respectively. The keywords are followed by a list with the chosen variables. The variables are within quotation marks and separated by commas. If the list with the variables has to be divided up into several lines, this should be done after a comma and not within the variable name.

MATRIX

Type: text Length: 20 Multiline: No

The name of the matrix. Is suggested as file name when the file is fetched. (new length 2008).

STUB

Type: text Length: 256 characters per value Multiline: Yes

At least one of the keywords STUB or HEADING must be included. Usually both are included, as you choose one or several variables for the stub and the heading, respectively. The keywords are followed by a list with the chosen variables. The variables are within quotation marks and separated by commas. If the list with the variables has to be divided up into several lines, this should be done after a comma and not within the variable name.

SUBJECT-AREA

Type: text Length: 256 characters Multiline: No

The name of the subject area in plain text, as shown in the menu "Select subject area". The text must not exceed 256 characters (new length in 2008).

SUBJECT-CODE

Type: text Length: 20 characters Multiline: No

Subject area code. It is used to create files with tables available in PC-Axis. The text must not exceed 20 characters (new length 2008).

TITLE

Type: text Length: Lines of 256 characters Multiline: Yes

The title of the table, reflecting its contents and variables. Is not needed if the file contains DESCRIPTION, in that case the text belonging to description is used. How the heading of a table will look depends on which variables the user chooses; the heading thus created will then be saved as TITLE. (Compare also DESCRIPTIONDEFAULT) If the file contains DESCRIPTION the text for this key word will be used.

UNITS

Type: text Length: 256 characters Multiline: No

Unit text, e.g. ton, index. The text must not exceed 256 characters (new length 2008). Compare UNITS for CONTVARIABLE. When there is a CONTVARIABLE the keyword UNITS takes an index and is repeated for every value for the contents variable. UNITS("value")="tons".

VALUES

Type: text Length: 256 characters per value Multiline: Yes

The keyword VALUES occurs once for each variable in the table, and is followed by the variable name in parentheses, within quotation marks. The values will be in the same order as in the stub and heading, respectively. They are within quotation marks and separated by commas. Each value name must not exceed 256 characters. If the values have to be divided up into several lines, this should be done after a comma and not within the value name. See also the keyword TIMEVAL below.

Optional keywords

AGGREGALLOWED

Type: {YES,NO} Length: N/A Multiline: No

If the contents of the table cannot be aggregated, contains for instance index and average, the keyword AGGREGALLOWED=NO; is used to stop the user from making a sum. If the keyword is missing aggregations are allowed

ATTRIBUTE-ID

Type: text Length: 256 characters per value Multiline: Yes

This keyword list the identities of all attributes. Indirectly it also specifies the number of attributes. Each attribute id is separate with a comma sign and it is written within quotes. The ATTRIBUTE-ID keyword must be written before the ATTRIBUTES keyword and must exist if the ATTRIBUTES keyword is present. The ATTRIBUTE-ID keyword is multilingual, that is you do not have to specify it for each language.

Example

ATTRIBUTE-ID="ObsStatus","ObsConf";

ATTRIBUTE-TEXT

Type: text Length: Lines of 256 characters Multiline: Yes

This keyword specifies a textual representation of the attribute for presentational purpose and it is optional. Each attribute text is separate with a comma sign and written within quotes. The ATTRIBUTE-TEXT keyword must occur after the ATTRIBUTE-ID and it is also language dependent. The order of the texts should be the same as for the order of the ids.

Example

ATTRIBUTE-TEXT="Observation status","Observation confidence";
ATTRIBUTE-TEXT["sv"]="Status","Tillit";

ATTRIBUTES

Type: text Length: Lines of 256 characters Multiline: Yes

This keyword specifies all attributes for as single observation value. Each attribute value is separated with a comma sign and written within quotes. The attribute values must appear in the same order as they appeared in the ATTRIBUTE-ID keyword. The ATTRIBUTES keyword must be written after ATTRIBUTE-ID and the CODES keyword. Which observation it refers to is given by the key which is written within parenthesis after the keyword and before the equal sign. The format of the key is one value code for each variable in the order indicated by STUB and HEADING starting with the variables in the STUB. If the attribute key is missing then this will indicate the default values for the attributes for all measures that do not have any ATTRIBUTES specified. This is also referred to as the default attributes. The default attributes should always be specified in order to save space and unnecessary repetition.

Example

ATTRIBUTES("01","A01","2005")="A","P";
ATTRIBUTES="A","F";

AUTOPEN

Type: {YES,NO} Length: N/A Multiline: No Default value: NO

If the file is published on the Internet and the user selects a number of variables and values it is possible to remove the windows “Select variables and values” in PC-Axis and instead show the complete table in PC-Axis when the file is downloaded. AUTOPEN=YES;.

AXIS-VERSION

Type: text Length: 20 characters Multiline: No

Version number for PC-Axis (new text length 2008). Is read and saved but otherwise not used.

BASEPERIOD

Type: text Length: Lines of 256 characters Multiline: Yes

Base period for, for instance index series. Is shown with the footnote. If there is a contents variable the keyword is repeated for each value of the contents variable.

CELLNOTE

Type: text Length: Lines of 256 characters Multiline: Yes

Footnote for a single cell or a group of cells. Which cell it refers to is given by values and variables. If a value is given as * the note refers to all values for that variable. Only one value can be given for each variable. The values are given in the variable order indicated by STUB and HEADING, starting with STUB.

Example

CELLNOTE("*","*","Örebro", "1995")="Lekebergs kommun has been excluded from Örebro";

CELLNOTEX

Type: text Length: Lines of 256 characters Multiline: Yes

As CELLNOTE but shown mandatory as for NOTEX.

CFPRICES

Type: {C,F} Length: N/A Multiline: No

Indicates if data is in current or fixed prices. C is used for Current and F for Fixed prices. Quotation marks must be used. CFPRICES="C" or CFPRICES("value")="C".

CHARSET

Type: text Length: 20 characters Multiline: No

CHARSET=”ANSI”; indicates that the texts in the file are written in Windows format. If the keyword is missing it means that the texts in the file are in DOS format. They will be translated by PC-Axis to Windows. This keyword must appear in the beginning of the file before any texts that can include characters outside A-Z, 0-9.

CODEPAGE

Type: text Length: 20 characters Multiline: No

Is used when creating XML format to get correct characters. Default iso8859-1. Max 20 chars.

CODES

Type: text Length: Lines of 256 characters Multiline: Yes

The key word CODES is used if a variable exists both in code and plain text. The codes are written in the same way as VALUES. Not more than 256 characters.

CONFIDENTIAL

Type: integer Length: 20 characters Multiline: No Default value: 0

Possibility to do some manipulation with the data in the data part of the file. Is only suitable if the user cannot download the total file since the data can be read in any editor. Max 20 chars. The only other value that it can take is 1 and that uses the simple disclosure control. Which is that it displays values of 1 to be 0 and values of 2 to be 3.

CONTACT

Type: text Length: Lines of 256 characters Multiline: Yes

States the person who can give information about the statistics. Is written in the form name, organization, telephone, fax, e-mail. Several persons can be stated in the same text string and are then divided by the #-sign. Is shown with the footnote. If there is a contents variable the keyword is repeated for each value CONTACT("value")="xx".

Example

CONTACT="Maria Svensson, SCB, +4619176800, +4619176900,
maria.svensson@scb.se";

CONTVARIABLE

Type: text Length: 256 characters Multiline: Yes

This is used to indicate that the table has two or more different contents. For instance the contents Import and the contents Export. The variable name must also be found either as STUB or HEADING. When a content variable exists a number of keywords will be indexed: DAYADJ, SEASADJ, STOCKFA, UNITS, CONTACT, LASTUPDATED, REFPERIOD, BASEPERIOD, CFPRICES. The keyword CONTVARIABLE must proceed the first keyword that will be indexed.

COPYRIGHT

Type: text Length: 20 characters Multiline: No

Copyright is given as YES or NO. If COPYRIGHT=YES the copyright refers to the organization given in SOURCE. Is shown together with footnotes.

CREATION-DATE

Type: text Length: 256 characters Multiline: Yes

Date when file was created. Written in format CCYYMMDD hh:mm, e.g. ”19960612 14:20”. Is shown together with footnotes.

DATABASE

Type: text Length: Lines of 256 characters Multiline: Yes

The name of the database from where the statistics is retrieved. Is shown with the footnote.

DATANOTE

Type: text Length: 20 characters Multiline: Yes

The DATANOTE keyword is used to indicate that a note exists for a certain element of the statistical cube If no sub key is specified like below the data note * is valid for the hole table. DATANOTE="*"; If there is just one part sub key like below. Then the sub key must refer to a variable and the data note is for that variable. DATANOTE("VARIABLE")="*"; The example above states that the data note * should be displayed for variable VARIABLE. If the sub key is divided into two parts. Then the first part of the sub key refers to a variable and the second part refers to a value for that variable. DATANOTE("VARIABLE","VALUE")="*"; The example above states that the data note * should be displayed for value VALUE for variable VARIABLE. The keyword is multilingual so different data notes can be used for different languages. It is recommended to add the data note to the note text. E.g. if you have a note on the time period 2010 it could look something like this in the PX file.

DATANOTE("tid","2010")="*";
VALUENOTE("tid","2010")="* Preliminary results";

DATANOTECELL

Type: text Length: 20 characters Multiline: No This keyword needs information for which value for each variable it applies and the symbol is presented together with the figure. CODES are used in the parentheses to indicate the right value. * means all values for the variable.

Example

DATANOTECELL("*", "20", "*", "BE0101F2", "*")="Ae";

DATANOTESUM

Type: text Length: 20 characters Multiline: No

This keyword gives information which symbol is presented together with the figure after a sum has been made consisting of differing datanote symbols.

DATASYMBOL1, DATASYMBOL2, DATASYMBOL3, DATASYMBOL4, DATASYMBOL5, DATASYMBOL6

Type: text Length: 20 characters Multiline: No

This if used to indicate how a stored “.” is to be presented in a table.

DATASYMBOLNIL

Type: text Length: 20 characters Multiline: No

This is used to indicate how a stored “-” is to be presented in a table. The – indicates that the figure is absolutely nil.

DATASYMBOLSUM

Type: text Length: 20 characters Multiline: No

This if used to indicate how a sum of differing numbers of dots will be shown. The sum is stored as “…….”.

DAYADJ

Type: {YES, NO} Length: N/A Multiline: No

DAYADJ=YES means that data is adjusted e.g. to take into account the number of working days. Default is DAYADJ=NO or DAYADJ("value")=NO.

DEFAULT-GRAPH

Type: integer Length: 1-10 Multiline: No

This keyword is read and saved in the PX-file but not shown in PC-axis.

DESCRIPTIONDEFAULT

Type: {YES, NO} Length: N/A Multiline: No

For some languages it is difficult to build a table title dynamically. The keyword DESCRIPTIONDEFAULT=YES; means that the text after keyword Description will be used as title for the table.

DIRECTORY-PATH

Type: text Length: 256 characters Multiline: Yes

This keyword is read and saved in the PX-file but not shown in PC-axis.

DOMAIN

Type: text Length: 256 characters Multiline: Yes

Can occur once for each variable. Is used to determine which value sets are of interest, and thus which aggregation lists can be used. (New text length 2008).

DOUBLECOLUMN

Type: {YES, NO} Length: N/A Multiline: No

This keyword is used to get code and text in separate columns for the specified variable DOUBLECOLUMN(”region”)=YES;. On screen it only has effect if the user selects presentation in matrix format.

ELIMINATION

Type: {YES, NO, text} Length: 256 characters Multiline: Yes

States if and how a variable may be eliminated in a table. If the key word is written as ELIMINATION("variable name")="value name" this value will be used as an elimination value if the user does not select the variable to the table. If the key word is written ELIMINATION("variable name")=YES this means that the variable will be eliminated by the summing up of all the values for that variable in the file.

FIRST-PUBLISHED

Type: {YES, NO, text} Length: 256 characters Multiline: Yes

The date when the data cube was first published in the format CCYYMMDD hh:mm.

Example

FIRST-PUBLISHED=20130224 20:55

HIERARCHIES

Type: text Length: Lines of 256 characters for each value Multiline: Yes

Example 1

HIERARCHIES(“Country”)="parent","parent":"child",...

Example 2

HIERARCHIES("Country")=”E25","E25":"E15","E15":"E12",
"E12":"AT","E12":"BE","E12":"FI","E12":"FR","E12":"DE",
"E12":"GR","E12":"IR","E12":"IT","E12":"LU","E12":"NL",
"E12":"PT","E12":"ES","E15":"DK","E15":"SW","E15":"UK",
"E25":"E10","E10":"CY","E10":"CZ","E10":"EE","E10":"HU",
"E10":"LV","E10":"LT","E10":"MT","E10":"PL","E10":"SK", "E10":"SI";

HIERARCHYLEVELS

Type: integer Length: N/A Multiline: No

Indicate the number of levels existing for a symmetrical tree.

Example

HIERARCHYLEVELS("Country")=4

HIERARCHYLEVELSOPEN

Type: integer Length: N/A Multiline: No

Indicate how the tree is to be displayed when the tree is shown for the first time

Example

HIERARCHYLEVELSOPEN("Country")=1

the first level is open.

HIERARCHYNAMES

Type: text Length: Lines of 256 characters per name Multiline: Yes

To give names to the levels in a symmetrical tree.

Example

HIERARCHYNAMES("Country")="NameOfLevel1","NameOfLevel2",..

INFO

Type: text Length: Lines of 256 characters Multiline: Yes

This keyword is read and saved in the PX-file but not shown in PC-axis.

INFOFILE

Type: text Length: Lines of 256 characters Multiline: Yes

Name of a file containing more information for the statistics. If the keyword exists in the file a new button is shown in the toolbar and the user can click for more information. Depending on the file type and to which program the type is associated the corresponding program is started and the information shown. (Word for .DOC, Notepad for .TXT etc).

KEYS

Type: {VALUES, CODES} Length: N/A Multiline: No

If this keyword is used it must occur as many times as there are variables in the stub. It contains the name of the variable and whether the key is taken from VALUES or CODES.

Example

KEYS("age")=VALUES;
KEYS("region")=CODES;

When it is used all data rows start with the Value/code of the variables in the stub. See Description of DATA.

LANGUAGE

Type: text Length: 2 characters Multiline: No The language used in the PC-Axis file (2 chars), sv for Swedish, en for English etc. Compare language codes for text files. If the keyword is used, the words for “and” and “by” are read from the text file of that language. Otherwise these words are read from the text file of the language in which the program is running.

LANGUAGES

Type: text Length: 2 characters per language Multiline: Yes

If more than one language in the PC-Axis file then all languages are mentioned here (2 chars each)

Example

LANGUAGES="en","sv";

LAST-UPDATED

Type: text Length: 256 characters Multiline: No

Date and time for latest update format CCYYMMDD hh:mm. Example ”19960528 11:35”. Is also used in Aremos file format. The date is not updated in PC-AXIS when changes are made to the table. If there is a contents variable it is written as LAST-UPDATED("value")= 19990318 18:12

LINK

Type: text Length: 256 characters Multiline: Yes

This keyword is read and saved in the PX-file but not shown in PC-axis.

MAP

Type: text Length: 256 characters Multiline: Yes

Used for a geographic variable for which maps can be made (new text length 2008).

Example

MAP("region")="Sweden_municipality";

META-ID

Type: text Length: 256 Multiline: No

The META-ID keyword is used to reference a external meta information about a table, variable or value. I there is no subkey specified the META-ID refers to the cube META-ID="RT:12"; If there is just one part sub key like below. Then the sub key must refer to a variable and the data note is for that variable. META-ID("VARIABLE")="V:167"; The example above states that the reference id for variable VARIABLE is V:167. If the sub key is divided into two parts. Then the first part of the sub key refers to a variable and the second part refers to a value for that variable. META-ID("VARIABLE","VALUE")=VAL:232; The example above states that the reference id for value VALUE for variable VARIABLE is VAL:232.

NEXT-UPDATE

Type: text Length: 256 characters Multiline: Yes

This keyword is read and saved in the PX-file but not shown in PC-axis.

NOTE

Type: text Length: Lines of 256 characters Multiline: Yes

Contains a footnote which is showed in the statistical databases if the user demands it. In PC-AXIS it is shown if the user presses F7. The footnote may either refer to the entire table or to a table variable. In the latter case the key word must be followed by the variable name in parentheses.

NOTEX

Type: text Length: Lines of 256 characters Multiline: Yes

Contains a note which is always shown in the statistical databases. In PCAXIS the note is shown automatically, before the table is presented on the screen. The note may either refer to the entire table or to a table variable. In the latter case the key word should be followed by the variable name in parentheses.

OFFICIAL-STATISTICS

Type: {YES, NO} Length: N/A Multiline: No

Indicates if the data table is included in the official statistics of the organization.

PARTITIONED

Type: complex text Length: N/A Multiline: Yes This is used to partition a variable into levels, for instance

PARTITIONED("region")="municipality",1,4;
PARTITIONED("region")="subarea",5;

states that the first four positions for the regional values contain the municipality code, and that the subarea code starts in position 5. Thus the values for the variable region after the key word VALUES must be written in code, not plain text. Max 3 levels can be used, each gives start position and length except for the last level where length is implied as rest of the code.

PRECISION

Type: integer Length: N/A Multiline: No Can occur for single values. Determines that the value shall be presented with a number of decimals that differs from the keyword SHOWDECIMALS. Is to be written as PRECISION("variable name","value name")=n where n is a figure between 1 and 6. The number of decimals for precision must be higher than the number of decimals for SHOWDECIMALS to have any effect.

PRESTEXT

Type: integer Length: N/A Multiline: No

From PX file format version 2010 and later the definition of PRESTEXT is changed the prior versions defines it as: States if texts or codes are shown for the keyword VALUES. Normally a file is created so that texts are found after the keyword VALUES and codes after the keyword CODES. This is equivalent to PRESTEXT("variable name")=1; which is the default. If a user changes presentation from texts to codes and saves the file the value codes will be saved after the keyword VALUES and value texts after the keyword CODES. In this case the keyword PRESTEXT is written as PRESTEXT("variable name")=0. This enables PC-Axis to know if it is necessary to switch to codes if aggregation is selected. The user can also decide to show both codes and texts for a value and in this case the keyword is saved as PRESTEXT("variable name")=2 or PRESTEXT("variable name")=3. PRESTEXT becomes 2 if it originally was 1 and 3 if it originally was 0. In the new definition the PRESTEXT is that it is an integer value between 0- 3 that describes how the value should be presented in a user interface. The values text should always be defined by the VALUES keyword and the codes should always be defined by the CODES keyword. - 0 - Only the value code should be displayed. - 1 - Only the value text should be displayed. - 2 - Both code and value should be displayed and the order should be the code and then the value text. - 3 - Both code and value should be displayed and the order should be the value text then the value code.

Example

VALUES("region")="Albania","Austria",…
…
CODES("region")="AL","AT",…
…
PRESTEXT("region")=2
Since the PRESTEXT is set to 2 the values would be presented as
AL Albania
AT Austria
…

PX-SERVER

Type: text Length: 256 characters Multiline: Yes

This keyword is read and saved in the PX-file but not shown in PC-axis.

REFPERIOD

Type: text Length: 256 characters Multiline: Yes

Text with information on the exact period for the statistics. Is shown with the footnote. If there is a contents variable the keyword is repeated for each value of the contents variable.

ROUNDING

Type: integer Length: 0-1 Multiline: Yes

If rounding is missing then rounding will be done in accordance with information in the pcax2000.ini file. If the keyword is missing then the decimal rule in pcax2000.ini is used. - Rounding = 0 means that .5 is rounded up if the previous figure is odd, 1.5 becomes 2. - Rounding = 1 means that .5 is always rounded up. 0.5 becomes 1 and 1.5 becomes 2.

SEASADJ

Type: {YES, NO} Length: N/A Multiline: No

SEASADJ=YES means that data is seasonally adjusted. Default is SEASADJ=NO or SEASADJ("value")=NO.

SHOWDECIMALS

Type: integer Length: N/A Multiline: No

The number of decimals to be shown in the table, 0-6. Must be the same or smaller than the number stored as indicated by the keyword DECIMALS. If SHOWDECIMALS is not stated in the file the number stated by DECIMALS will be used.

SOURCE

Type: text Length: 256 characters Multiline: Yes

States the organization which is responsible for the statistics. Is shown with the footnote.

STOCKFA

Type: {S, F, A} Length: N/A Multiline: No

Indicates if data is stock, flow or average. The used characters S (stock), F (flow) and A (average) must be within quotation marks. STOCKFA="S" or STOCKFA("value")="S".

SURVEY

Type: text Length: 256 characters Multiline: Yes

This keyword is read and saved in the PX-file but not shown in PC-axis. Is shown on information screen in PX-web if installation parameter true.

SYNONYMS

Type: text Length: 256 characters Multiline: Yes

This is used for search in PX-Web. If the table concerns CPI the synonym can be “Inflation”. Several words can be included within “ “ and the total line can be max 256 chars.

TABLEID

Type: text Length: 256 characters Multiline: Yes A text that is the identity of the table.

TIMEVAL

Type: text Length: N/A Multiline: Yes

New keyword in version 1.5 to enable the use of time series. After the keyword is the name of the time variable given, e.g. TIMEVAL(”time”). TLIST gives information on timescale and timeperiods. The time periods must be consecutive. If the table contains the time periods 1994, 1995, 1996 TIMEVAL can be written in this way:

TIMEVAL(”time”)=TLIST(A1, ”1994”-”1996”);

or

TIMEVAL(”time”)=TLIST(A1), ”1994”, ”1995”,"1996”;

The following formats are used: - A1 for annual statistics written as CCYY (C for century, Y for year) - H1 for halfyear in format CCYYH, where H is 1 or 2 - Q1 for quarterly data written CCYYQ, where Q is 1 - 4. - M1 for monthly statistics written CCYYMM - W1 for weekly data written CCYYWW

TIMEVAL is used in PC-AXIS version 1.7 when converting to Aremos and Gesmes/Ecoser format. It is also used for start and stop time in list of contents in HTML-format. The variable time can be used both after the keyword VALUES and after TIMEVAL. If both exist TIMEVAL must be placed after VALUES. If only TIMEVAL exists VALUES for time are created as follows: Time scale time periods shown as

A1 1995, 1996, 1997
H1 1995H1, 1995H2, 1996H1
Q1 1995Q1, 1995Q2, 1995Q3
M1 1995M01, 1995M02, 1995M03
W1 1995W01, 1995W02, 1005W03

UPDATE-FREQUENCY

Type: text Length: 256 characters Multiline: Yes

This keyword is read and saved in the PX-file but not shown in PC-axis.

VALUENOTE

Type: text Length: Lines of 256 characters Multiline: Yes

A footnote for separate variable values. Should be written with the variable name and the value names in parentheses.

VALUENOTEX

Type: text Length: Lines of 256 characters Multiline: Yes

Mandatory footnote for single values for a variable. Is written with the variable name and the value name in parentheses. Is shown the same way as NOTEX.

VARIABLE-TYPE

Type: text Length: 256 characters Multiline: Yes

This keyword is read and saved in the PX-file but not shown in PC-Axis.

Recommended order of the keywords

Some keywords must be written before some other keywords e.g. STUB and HEADING must precede VALUES and CODES. Here follows a list of recommended order for how keywords should be written I PX-files though applications should not depend on that this order is used.

• CHARSET
• AXIS-VERSION
• CODEPAGE
• LANGUAGE
• LANGUAGES
• CREATION-DATE
• NEXT-UPDATE
• PX-SERVER
• DIRECTORY-PATH
• UPDATE-FREQUENCY
• TABLEID
• SYNONYMS
• DEFAULT-GRAPH
• DECIMALS
• SHOWDECIMALS
• ROUNDING
• MATRIX
• AGGREGALLOWED
• AUTOPEN
• SUBJECT
• CODE
• SUBJECT-AREA
• CONFIDENTIAL
• COPYRIGHT
• DESCRIPTION
• TITLE
• DESCRIPTIONDEFAULT
• CONTENTS
• UNITS
• STUB
• HEADING
• CONTVARIABLE
• VALUES
• TIMEVAL
• CODES
• DOUBLECOLUMN
• PRESTEXT
• DOMAIN
• VARIABLE-TYPE
• HIERARCHIES
• HIERARCHYLEVELS
• HIERARCHYLEVELSOPEN
• HIERARCHYNAMES
• MAP
• PARTITIONED
• ELIMINATION
• PRECISION
• LAST-UPDATED
• STOCKFA
• CFPRICES
• DAYADJ
• SEASADJ
• UNITS
• CONTACT
• REFPERIOD
• BASEPERIOD
• DATABASE
• SOURCE
• SURVEY
• LINK
• INFOFILE
• FIRST-PUBLISHED
• META-ID
• OFFICIAL-STATISTICS
• INFO
• NOTEX
• NOTE
• VALUENOTEX
• VALUENOTE
• CELLNOTEX
• CELLNOTE
• DATASYMBOL1
• DATASYMBOL2
• DATASYMBOL3
• DATASYMBOL4
• DATASYMBOL5
• DATASYMBOL6
• DATASYMBOLSUM
• DATASYMBOLNIL
• DATANOTECELL
• DATANOTESUM
• DATANOTE
• KEYS
• ATTRIBUTE-ID
• ATTRIBUTE-TEXT
• ATTRIBUTES
• PRECISION
• DATA