C# Class SIL.FieldWorks.UnicodeCharEditor.PUAInstaller

Adds a number to a hex string

Adds/"deletes" a given puaCharacter to the given TextWriter Will write a DerivedBidiClass.txt style line.

Create the "original" (backup) copy of the file to be modified, if it doesn't already exist.

Replace the file extension with 'newExtension'.

This method will create the backup of the original input file.

This method appends 'nameSplice' to a file 'inputFilespec'.

This method will create the temporary work file copy of the original input file.

Reads the Property value from the given line of a UCD text file. (For example, if the text file is DerivedBidiClass.txt, this reads the bidi class value from the given line of a DerivedBidiClass.txt file.)

Installs the PUA characters (PUADefinitions) from the given xml file We do this by: 1. Maps the XML file to a list of PUACharacter objects 2. Sorts the PUA characters 3. Opens UnicodeDataOverrides.txt for reading and writing 4. Inserts the PUA characters via their codepoints 5. Regenerate nfc.txt and nfkc.txt, the input files for gennorm2, which generates the ICU custom normalization files. 6. Run "gennorm2" to create the actual binary files used by the ICU normalization functions.

Given a regular file name, this method returns the file name that is used for storing the 'origional' version of the file.

Remove the backup files that were created to restore original files if there were a problem. If this method is called, there was no problem. We're just removing the backup file - much like a temp file at this point.

Copies the backup files over the source files and then deletes the backup files.

This runs genprops and gennames in order to use UnicodeData.txt, DerivedBidiClass.txt, as well as other *.txt files in unidata to create icuprefix/uprops.icu and other binary data files.

Checks to make sure the given codeIndex is within the current bidi section. Performs bounds checking as well.

Add a file to the list of temporary files.

Checks whether the IPuaCharacter needs to be added to the lists, and adds if necessary.

Create a copy of the backup and source file names for future use.

Check for locked ICU files and return

This method will create the temporary work file copy of the original input file.

Retrieves the given field from the given UnicodeData.txt line

returns true if the line is not just a comment i.e if it is of either of the following forms ------- ; ------ # ------ ------- ; ------- NOT ----- # ----- ; ------

Given a line containing a valid code or code range, increments the count to include the code or code range. Uses XXXX..YYYY style range.

Inserts the given PUADefinitions (any Unicode character) into the UnicodeData.txt file. This accounts for all the cases of inserting into the "first/last" blocks. That is, it will split the blocks into two or move the first and last tags to allow a codepoint to be inserted correctly. Also, this accounts for Hexadecimal strings that are within the unicode range, not just four digit unicode files. Assumptions made about the format The codepoints are in order There first last block will always have no space between the word first and the following ">" No other data entries contain the word first followed by a ">" There will always be a "last" on the line directly after a "first".

Pseudocode for inserting lines: if the unicodePoint is a first tag Get first and last uncodePoint range Stick into array all the xmlPoints that fit within the uncodePoint range Look at the next xmlPoint if there are any call WriteCodepointBlock subroutine else if the unicodePoint is greater than the last point but less than or equal to "the xmlPoint" insert the missing line or replace the line look at the next xmlPoint else do nothing except write the line

Prints a message to the console when storing a Unicode character.

Makes three debug files for the DerivedBidiClass.txt: A: Saves the "in-between" state (after deleting, before adding) B: Saves a list of what we are adding and deleting. C: Saves an actual file that doesn't add or delete anything. This will update the numbers, but won't do anything else

This function will add or remove the given PUA characters from the given DerivedBidiClass.txt file by either inserting or not inserting them as necessary as it reads through the file in a single pass from tr to tw.

Assumptions The like Bidi values are grouped together Non Assumptions: That comments add any information to the file. We don't use comments for parsing. That "blank" lines appear only between different bidi value sections. That the comments should be in the format: # field2 [length of range] Name_of_First_Character..Name_of_Last_Charachter (If it is not, we'll just ignore.

Read to the end of a given section of matching bidi class values. Passes everything read through to the writer. If this finds a section count comment, it will replace it with the current count.

Delete all files in the list of temporary files.

Subtracts two hex numbers and returns an integer value of their difference. i.e. returns an int representing hex-hex2

Updates a UCD style file as necessary. so that the entries match the ones we just inserted into UnicodeData.txt.

A UCD file is a "Unicode Character Database" text file. The specific documentation can be found at: http://www.unicode.org/Public/UNIDATA/UCD.html#UCD_File_Format

Write a codepoint block, inserting the necessary codepoints properly.

Writes a block representing a range given first and last. If first is not before last, it will do the appropriate printing.

Write a codepoint block, inserting the necessary codepoints properly.

Writes a block representing a range given first and last. If first is not before last, it will do the appropriate printing.

Writes a UnicodeData.txt style line including comments.