Technical 10: Text Conversion Attributes

2003 TextMaestro Technologies

	Technical
1.	Introduction
2.	Main Toolbar
3.	Session Toolbar
4.	FTP Toolbar
5.	File Attributes
6.	Project Manager
7.	Account Manager
8.	Browse Remote System
9.	Text Conversion Dialog
10.	Text Conversion Attributes
11.	Text Conversion Libraries
12.	Text Conversion Rules
13.	Text Conversion Batch Mode
14.	Text Conversion Command Line
15.	Report
16.	Diff-and-Merge
17.	Cross Listings
18.	Program Options
19.	Save Program Data

10. Text Conversion Attributes

When you click on Text Conversion button from Main toolbar to open the Interactive Text Conversion dialog, a secondary dialog appears as shown below:

Usage of the attributes deeply affect the behavior of the text conversion.

Although their roles can be sometimes subtle, they are indeed powerful.

A brief description follows.

The technical aspect of Attributes needs some detailed discussions. Key concept of text conversion feature of TextMaestro spans over this section, Conversion Schemes and libraries, Batch Conversion, Quick Convert Help and a series of Examples. It is strongly recommended that you read these related topics to understand the heart of the matter.

Recall, above Attributes work in conjunction with the Interactive Text Conversion Window.

Find Text

Supply what to be found in the Input text. Without checking Use wild-card, this is equivalent to regular find-and-replacement in common text editors.

Replace Text

Supply text that should replace the found text.

Match case

Check this if the search should be case sensitive. Matching case enhances the conversion performance. Ignoring case might make a large conversion process surprisingly slow. It is recommended that you select Match case as much as possible.

Use wild-card for Find

Click on this to enable the scheme of matching wild-card for Find and Replace text. When Use wild-card is checked, Find and Replace scheme takes a whole different meaning, and thereby a new set of controls expose where you have really a good grasp to drive your conversion scheme.

Example:

Consider this set up:

I.e., consider ab*cd*ef for Find as shown above, where Use wild-card in Find is checked. This will result in the following search for input text:

Search for ab.
If found continue searching until cd is found, and whatever you find on the way, mark that as wild card argument one.
If cd is found, continue searching until ef is found, and whatever you find on the way, mark that as wild card argument two.
If any of ab, bc or ef is not found (in that order) then search result is unsuccessful.

Use wild-card for Replace

Here are the pictures:

When Use wild-card is enabled for Replace, the wild cards arguments constructed from using wild cards in Find play a vital role. Continuing with the above example, consider a Replace pq*rs*tu and consider the following input text:

The texts I found are

ab-is-argument-onecd-then-argument-twoef

and some more.

As noted from color highlighting, first * stands for -is-argument-one, and the second *, for -then-argument-two. Thus, with ab*cd*ef as Find, and pq*rs*tu as Replace, the following output will be generated from the above input:

The texts I found are

pq-is-argument-oners-then-argument-twotu

and some more.

See below.

There are various rules and qualifiers involving wild-card usage. See Quick Convert Help to get started. See Example 14 for a demo on wild-card usage.

When Use wild-card for Find is checked, the Wild-Card Manager button is enabled, clicking on which, you see the following dialog:

Here user can define his own set of type-cards. In gist, wild-card denotes "anything", and type-card denotes one or more of the character set assigned to it. What you see above is the default setting. Every Find and Replace library has its own Wild-Card Manager, thereby, its own set of type-cards. Type cards are extremely useful when you want to parse various log files where you want to identify a certain sequence of characters as a set, and then grab them by the corresponding type-card during parsing.

Knobs...

If you click on hyperlink Knobs..., this dialog will appear.

If you click on the hyper-links on the right side, marked by , , and , corresponding Quick Convert Help slides will appear.

Below is the description of the four knobs.

Per-line search: This works in conjunction with Use wild-card for Find. When the parsing builds a wild-card argument, the search might span over multiple lines. If this control is checked, the search is forced to discontinue when a line break is encountered inside a constructed wild-card argument, and continue for a new search from the current position. This is useful when you want to force the wild-card construction remain within a line.

Compact construct: This works in conjunction with Use wild-card for Find. Use this to advance search handle to determine if there can be a match within a match in order to make Find of minimal span. This is useful when you want to grab a block of text.

Filter processed text: This works in conjunction with Use wild-card for Find. Use this when you want to keep only the found and replaced text and discard everything else. This is useful when you are interested in certain type of text blocks in a large dump or source code. For example, class*}; for Find and class*}; for Replace, and having Filter processed text checked will display all class declarations from a set of .h files (with some set up under Batch Conversion).

Checked	Filter processed text
Find:	class*};
Replace:	class*};

Grabs all class declarations.

Delimiter parse: This works in conjunction with Use wild-card for Find. Use this when you want to grab a block of text delimited by a pair of delimiters listed in the Delimiter List, accessed by the button found next to the Delimiter parse check box. In such case, if Find string is found in a block of text, identified by the delimiter, then Replace string is used to replace the whole text block (not only the Find string.) Details are available in 12.b. Conversion Rules - Find And Replace | Knobs | Delimiter parse.

Top