C# classes for handling javascript Source Maps (.map files for in-browser javascript debugging)

Copied and converted from "dart2js.dart" source code (Dart language SDK).

This example demostrates the use of map files over a fictional language "UPPECARSE-JAVASCRIPT" where the source files are made of Javascript written all upercase.

The test program "compiles" (merges) source1.txt + source2.txt and produces myapp.js and myapp.js.map as result.

To run the example:

• clone this repo
• open in visual studio
• run the TestExample project. This will create myapp.js and myapp.js.map files
• place some breakpoints in source1.txt or source2.txt
• run the Website project in visual studio with IE browser
• see the debugging in action

• Source maps works better if the granularity is set at token level (punctuation included)
• If a coarse granularity is used, the highlight selection will span over, reaching the start of the next map
• A granularity at character level is possible too, e.g. mapping every single character, this seem to work and to be handled ok by VS and the browsers
• If there are more map points than necessary, both browsers and VS will "intersect" them with legal points in the javascript code
• VS uses maps to make debugging at istruction level, while browser work at "start of line" level
• If there are two istructions on the same line, VS will handle it, browsers instead will consider just one breakpoint for the line
• The names specified in the maps are used only for callstack tracing, so only names for function calls should be specified
• Variable names are useless, Visual Studio doesn't recognize them (hovering the mouse over the variable name does not produce the tooltip)