Is it time for the Lazarus IDE community to adopt an interpreter? lazarus / 8-Other ---------------------------------------------------------------------- vfclists · 2013-04-19 19:44 Is it time for the Lazarus IDE community to adopt an interpreter? Is it time for an interpreter to gain greater prominence in Lazarus and Pascal usage, get close to center stage? Adopting one may be better than creating one from scratch. One of the reasons why languages like Javascript, Ruby, Lua, Python (I have omitted PHP ) have gained prominence over the years is that being interpreters there is no need to go through the compile, link cycle to get results, especially in this web-oriented era. With Pascal even the simplest utility scripts need to be compiled to be useful. When there are bugs in them unless you have the whole development setup available which in Lazarus's case can be very difficult for Linux users in particular you can hardly fix them in good time. I am not familiar with implementing compilers or interpreters, but isn't time for an interpreter to gain better coding and debugging support within the IDE? At least if the benefits of using one proves itself the idea will gain momentum and contributions will come in. ---------------------------------------------------------------------- marcov · 2013-04-19 20:06 Changing the roots of the project is not really sane. If there is interest in such a thing, it is better reserved for a separate project, since trying to keep both compatibility with the current compiler based accomplishments and the interpreter change would only result in an unhappy marriage of convenience, where both partners hamper each other. ---------------------------------------------------------------------- macmike · 2013-04-19 23:54 Using http://wiki.lazarus.freepascal.org/InstantFPC basically gives you what you want doesn't it? ---------------------------------------------------------------------- Leledumbo · 2013-04-20 05:02 QuoteI am not familiar with implementing compilers or interpretersNow you should try implementing, and you'll see the difficulty. Some really modern (born 2006 or later) compilers have architecture that makes it possible to be an interpreter as well, but FPC is not designed as such and it would be difficult to refactor the architecture to support that. Anyway, the compile-link time is quite negligible for most programs and it can reduce development time by catching errors at compile time. Many interpreters don't really evaluate everything so runtime errors are more likely to happen and it increases development time considerably.