TextMate (Ruby) Tricks

John Muchow of Mac Developer Tips recently posted a screencast showing a neat way to debug your Ruby code.

Speaking of the Ruby bundle, Ciarán Walsh has a post about the design philosophy behind the mnemonics of the Ruby bundle and a few highlights.

And while there seems to be a preference for writing TextMate commands in Ruby, anything that you can run from a terminal (shell) can be used. A step-by-step tutorial about how to write TextMate commands in PHP is available at Ciarán’s blog. I pushed him to write this one, as I believe there are a lot of people who work with PHP in TextMate that never got around to writing custom commands in the false belief that they would have to learn elisp or some similarly obscure language :)

27 Comments

  1. 28 Apr 2008 | # p4bl0 wrote…

    That's interresting :-).

    But something far more insterresting is how is the development of TextMate 2.0 ? When will we have news about it ? I can't wait to know and see it ! :-p

  2. 28 Apr 2008 | # Henrik N wrote…

  3. 28 Apr 2008 | # David Madden wrote…

    Yeh, what he said!!

  4. 28 Apr 2008 | # Trev wrote…

    Boo!!! come on dangle some tasty TM2 carrots for us Allan

  5. 28 Apr 2008 | # Jakub Nešetřil wrote…

    Yeah, I have to say I'm a bit concerned about TextMate. I've been waiting for TM2 for over two years now. It's still a great editor, but it's on the path of becoming the IE6 of text editors. Please, don't let it die for all the fame (like BBEdit did).

    I've made TextMate the default editor in my company, directly influencing at least 15 purchases – and I am struggling to keep recommending it because of this.

    The lack of visibility into future plans for TextMate is distressing – and shows well one of the weaknesses of proprietary software. If TextMate was OSS, the wave of popularity would never allow it to stagnate so much for so long.

    That's not to say I want to criticize Allan for keeping it closed. I just think if you're a one-man-show project, you need to work SO MUCH HARDER on transparency – to keep the user community in the loop. Saying "sign up TM2 next up to Duke Nukem Forever" doesn't really help at all… :(

  6. 28 Apr 2008 | # Jonno wrote…

  7. 28 Apr 2008 | # Graham wrote…

    Jakub – I'm struggling to see why you should be concerned, given the quality of the product and the obvious attention that Allan is giving to TextMate (at least, it's obvious if you follow the mailing list). I understand why he doesn't want to risk giving people the wrong impression of what will make it in to TM2. I, for one, am quite happy to wait.

    And as for recommending it – isn't it still the best editor around for most purposes? If you accept that development is still active (even though you can't see what's happening), why not continue to recommend it?

  8. 29 Apr 2008 | # oliver taylor wrote…

    Saying that TM1 has become "stagnant" during the development of TM2 is like saying Tiger development stagnated when Apple began developing Leopard. And the fact that Allan is taking longer that you'd like him to has no bearing whatever on his commitment to the product or the community.

  9. 30 Apr 2008 | # Fabien Freling wrote…

    Anyway, it's been a while since TM2 has been announced, and we are many to wait for some news about the development.

    Even if the development is still active, it looks like a dead product :/ And knowing the codename of TM2 doesn't quenched our thirst.

  10. 30 Apr 2008 | # Thomas Aylott wrote…

    Bah. Dead shmead.

    TM1 is getting new major features all the time due to us in the community.

    TM2 may technically still be vapor, but I trust Allan when he says that it's so well along that he's using TM2 to write TM2.

  11. 30 Apr 2008 | # Pierre-Olivier wrote…

    Maybe Allan is trying to mimic Apple's "We don't discuss future products" habit. This is a very bad idea when you don't have such an engineering force. Moreover, you let the door open to other products, maybe not as good as TM, but with a clearly-defined roadmap and frequent major updates.

  12. 30 Apr 2008 | # Jakub Nešetřil wrote…

    Graham, Oliver – TM1 is indeed still a very fine editor.

    It's the total silence – not for 2 month, not for 6 month, but for more then a year – that's disturbing me. I really don't know what to think. Is TM2 still coming, or has it been dropped entirely? Should I seriously place it next to Duke Nukem Forever?

    I think we are all reasonable people, developers by profession. Everybody will understand if Alan says "hey, TM2 is coming out in 4 month with this feature set" and 2 months later he'll say "guys, I'm not making it, progress has been slow, these features will get an axe" or "guys, I'm not making it, I need 2 more months". After all, that's what Apple did with Leopard – and guess what – we've got over it.

    So – once again – I'm not advocating for shippng TM2 here and now, I'm advocating for greater transparency. Roadmap, goals, features, plans… And if they change – life goes on…

    Unfortunately I cannot keep up with the mailing list volume. I do however subscribe to the bundle development list to scan the progress there. And from what I can see there, the activity in some core bundles of interest to me (Javascript+libraries, CSS, Python, Perl) is dying, too. And it's not because there's nothing left to improve – far from that. Maybe it's just that the current bundle development has reached it's peak complexity and needs more infrastructure support to go on – or it's the community getting weaker – I can't tell. I for one wonder, whether it's worth investing development time into it – since I can't see the future. :(

  13. 03 May 2008 | # Paul McCann wrote…

    Jakub,

    from my perspective Allan's reluctance to explain the what/when/why of TM2 is totally understandable. (Yes, it is coming (as per others' comments above).) Publicly disclosing the feature set and/or roadmap prior to release is only going to lead to an avalanche of queries and complaints and cheering and whining about what has been included, what has been left out ("but you indicated in late 2006 that…"), what's coming and when, frustration ("that's it, I'm moving to emacs/vim/jedit/bbedit/smultron/nedit/gedit/winedt_in_parallels"), complaints ("but every modern text editor has…") etc etc. Doubtless this outpouring will happen eventually, but from where I sit it's a damn good idea to wait until the fox is released before giving the hounds anything to bay about.

    It is certainly a bit unfashionable to effectively be told "trust me", but in essence that is what we have been doing since TextMate was first created. So far I would say that users' faith has been repaid many times over.

    Cheers, Paul

  14. 03 May 2008 | # Jacob Rus wrote…

    Jakub:

    I suppose the question is about whether we trust Allan. I certainly do. When he says he has put in the better part of a year of full-time work on TM 2 (it might even be more by now), I believe him. When he says he has mostly rewritten all the parts of TM from scratch, to put the whole project on a sounder footing, and allow new kinds of awesome customizations we couldn't imagine before, I believe him.

    Your concern is that:

    “If TextMate was OSS, the wave of popularity would never allow it to stagnate so much for so long.”

    But I think that this is the wrong way to look at things. If TM was OSS (i.e. with no paid developer), it would be impossible to do the kind of deep structural changes that Allan is attempting with TM 2, because of feature-creep in the core app, and lack of long-range vision. There is clearly a trade-off here, but given how carefully thought-out the tools TM 1 provides to build amazing stuff on top of are, I'm looking forward to seeing what Allan has learned from the experience of watching them in use for a few years.

    TM’s “stagnation” now allows it to jump way ahead once Allan has polished things up enough for release. I'll take temporary pain for the promise of a shiny future.

  15. 03 May 2008 | # Jacob Rus wrote…

    “Maybe it’s just that the current bundle development has reached it’s peak complexity and needs more infrastructure support to go on – or it’s the community getting weaker – I can’t tell.”

    I think there are several factors.

    1. Things are getting “good enough” to the point where fewer new (read: energized) users are dying to make fixes.
    2. Allan working full time on TM 2 stuff means he is putting less of his own time into TM 1 bundles, and his activity spurs quite a bit of the rest.
    3. Progress in the core app always drove up short-term progress in the bundles. It has been slow for quite a while, so not exerting that push.
    4. The promise of shiny new foundations in TM 2, on which to build cooler stuff than we can make now, has the community sitting back a bit.
  16. 03 May 2008 | # Jakub Nešetřil wrote…

    Paul, Jacob,

    thanks for your comments. Indeed, I look forward to TM 2 just all the rest of you – and I appreciate the geniality of TM 1 over and over. And Indeed, the issue seems to be "do we trust Allan". Maybe it's easier for those of you who know Allan better. For me, as much as I love TM 1, there's nothing saying Allan couldn't go crazy from all the fame and money and decided to retire early… I might be stretching things a bit, but I believe you get my point.

    Regarding the argument has been repeated many times of course – and above above by Paul:

    Publicly disclosing the feature set and/or roadmap prior to release is only going to lead to an avalanche of queries and complaints and cheering and whining about what has been included

    And my question is: "so what"? Lock yourself out while you're coding, ignore comments/whining. Or don't report on features at all. That doesn't prevent you from updating the community with some sort of status let's say… once a month.

    Even if it's just "Hi guys, I'm still heavy at work, progress has been good this month, I'm not going to tell you what features are in, but I think if things go well, I should wrap up in September".

  17. 04 May 2008 | # Allan Odgaard wrote…

    Jakub: Both for PR and competitive reasons it is a bad idea to blog about non-shipping unfinished software (something I learned a wee bit too late, otherwise I would never have mentioned 2.0 in public).

    Having a privately held company post monthly information about what the employes have been up to is damn near insane, and I am not sure how you think it would change anything. I mean, it’s not like it brings us any closer to whatever next product I may or may not decide to release.

    That said, I do feel I show some openness on the mailing list and IRC channel, which is the venue for my communication with the community.

    As for your “core bundles […] dying”, April had several hundred commits and I am seeing great things from the community like the revamped Dialog v2 plug-in and the interactive input injection library. So I don’t see lack of interest in building stuff for TextMate despite lack of major updates to the core application.

    I btw find it ironic that you dismiss Paul’s reasoning with the suggestion I just ignore the potential backfire, yet it seems you’d argue that the current whining about lack of full disclosure should not just be ignored ;)

  18. 07 May 2008 | # dave wrote…

    Allan: Personally i think that your mistake was the fact that you want to deliver tm2 as a free update. I would charge money for it because everyone would pay for a tm2 delivered this year instead of duke nukem forever dates :(, which in turn would bring more revenue, and in so you could actually hire another seasoned programmer, to help you drive tm2 development to a higher pace. It's been 1 year since the tm2 announcement that it will be available and really it does feel like vaporware, and this should not happen, because there is room for improvement and having a faq where you tell that you should put tm2 next to duke nukem forever is really a way of saying that tm2 might never come out, which in turn this might drive people to move to a product with a more clear vision for the future. I think you made a great product, i hope you will take this comment in a good sense, but you need to have tm2 a paid product, and hire more guys if you need to have this product have a real roadmap.

  19. 09 May 2008 | # jeremiah wrote…

    dave, your comments indicate to me that alan is right in saying he should not have ever mentioned tm2 in public… then we could all be spending our time more productively working on the projects we know we should be, rather than whining about this thing we're all looking forward to having not been realized yet – or defending the 'software is insanely better when quality thought and time are the primary development motivators' argument. sales driven development has lowered the bar for accepted software quality enough.

    i am happy to wait for the time it takes to produce tm2. it sounds like alan's learned a lot and is intent on producing something that will be around for a long time – and will require less time, once complete, to produce 'major' updates.

  20. 22 May 2008 | # eric wrote…

    tm2 will be worth waiting for and i am sure i am speaking for many content tm1 users…

  21. 29 May 2008 | # Devon Weller wrote…

    I use TextMate for many hours almost every day. And it works really well.

    The way I look at it is this: TM1 is a superb product. Allan did the Mac community a great service by creating it. If TM2 comes out, then that is just icing on the cake.

  22. 03 Jun 2008 | # martin wrote…

    Effort not compensated will die out. There is only so much in any reservoir.

    One guy spending a year or two developing commercial tool and giving it out free of charge? Why? If not money, what is the other half of the feeding loop?

    This, like the public announcement of TM2 (tossing TM to merciless dynamics of crowd's expectations) is threatening the future of your phenomenal tool, Allan. The e-editor for windows inspired by your product is evolving rapidly. Will it eventually port on Macs? Eat it's parent? :) That would be a nice irony. And sad fate.

    Public announcement cannot be taken back. But you can easily change your "give it out" decision. Every sane person will understand and agree. Those who will not, will pay anyway. Because the tool has a value. Dead simple. (Not to mention the love affair with TM your customers have going on.)

  23. 03 Jun 2008 | # ScottBruin wrote…

    I too am surprised that TM2 is a free upgrade. I know I would be willing to pay for it. However, Allan's finances are his own business so I trust he knows what he is doing.

  24. 10 Jun 2008 | # erydor wrote…

    I trust him: TM2 is coming soon. It only needs Leopard… Snow Leopard, though!!

  25. 16 Jun 2008 | # pumpkin wrote…

    maybe the slogan "the missing editor" makes more sense now :) I still love TM1, but it feels like it's stagnating :) I miss the bleeding edge updates and seeing what was going on in the changelog

  26. 08 Dec 2008 | # ruby textmate wrote…

    There's also the rubyamp which is very nice for textmate + ruby: http://code.leadmediapartners.com/ -=Roger

  27. 11 Jun 2009 | # angkasuwan wrote…

    just getting into ruby

Comments closed, you can use the mailing list for discussion.