Re: [Factor-talk] Git

2010-11-14 Thread Kyle Cordes
On Sun, Nov 14, 2010 at 5:55 PM, Shaping wrote: >> git branch > > displays > > (no branch) > master Possibly helpful: not-on-a-branch is a place you can land partway through a rebase / merge process. It can be painful to get your head around the git way of doing things, if you have already wor

[Factor-talk] Integrating Browser functionality into Listener: John Benediktsson's Syntax Highlighting

2010-11-14 Thread Shaping
Hi John/all. I still have some Git exercises and maintenance to do tonight, but I tripped over this http://planet.factorcode.org/ and want to see where it might lead. The syntax highlighting is interesting to me. I'm wondering whether we can change the Listener GUI into a color vocab browser tha

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Balazs Toth
It seems as if we swayed away from the main topic a wee bit here. Quite nice git tutorial forming though :) On Nov 15, 2010, at 3:24 AM, Jim mack wrote: > There is a subfolder in my work folder called .git This is what I think of > as the repository. When you do a git add or commit, it's goi

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Jim mack
On Sun, Nov 14, 2010 at 2:45 PM, Shaping wrote: > > > There's a lot going on, so context is difficult here, but I was trying to > go back in time before you made changes, and suggest when working on Factor > main code, you make changes *in a branch,* sync > > > > "sync" means to resolve conflict

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Jim mack
There is a subfolder in my work folder called .git This is what I think of as the repository. When you do a git add or commit, it's going between these two places. When you do a clone, push, pull, or fetch you're going to some remote place as well. On Sun, Nov 14, 2010 at 3:45 PM, Shaping wr

Re: [Factor-talk] Git GUI: clean start

2010-11-14 Thread Shaping
Okay, I flushed my Factor repo, saving off to the desktop (for now) two changed clean-branch files, and a few new work vocabs/exercises. I cloned the remote Factor repo: git://factorcode.org/git/factor.git. The resulting local repo clone is about 116 MB. Its .git subdirectory is about 64 MB. I

Re: [Factor-talk] Git: branch of latest fixes to web server; build fails

2010-11-14 Thread Shaping
The problem is that I left factor.exe running. The pull won't even go through if factor.exe is not available for linking. Shaping -- Centralized Desktop Delivery: Dell and VMware Reference Architecture Simplifying ent

Re: [Factor-talk] Git: branch of latest fixes to web server; build fails

2010-11-14 Thread Shaping
Pull incorporates changes from a remote repository into the current branch. What happens if you don't have a current branch? I just ran the factor.cmd, but I am on (no branch). Do I have to merge that stuff into master? The strangeness here is that I was never on master, which apparently is supp

Re: [Factor-talk] Git: branch of latest fixes to web server

2010-11-14 Thread Shaping
Factor.cmd can take "clean" or "latest". I want the bignum fix, which I see is in branch latest. I want the recent web server fix from Slava, but I don't see it. Can someone tell me what branch this is in, or what the fix is called? Shaping ---

Re: [Factor-talk] Git

2010-11-14 Thread Shaping
I may have broken that. I will reclone. I'm using both Git Bash and Bit GUI, the first to check the behavior of the second. Shaping -Original Message- From: Chris Double [mailto:chris.dou...@double.co.nz] Sent: 2010-November-14, 17:59 To: factor-talk@lists.sourceforge.net Subject: Re:

Re: [Factor-talk] Git

2010-11-14 Thread Shaping
> Some of the doc says that your default branch is "master". This is not > correct. I just did a "git status" and see "Not currently on a branch" What they mean is when you first clone a repository you get a default branch called 'master'. And this is the default checked out branch.

Re: [Factor-talk] Git

2010-11-14 Thread Chris Double
On Mon, Nov 15, 2010 at 12:55 PM, Shaping wrote: > > So Git GUI makes the master branch by default, but does not put you on it. > Why is that a good thing to do? I have no idea, sorry. I stick with the command line. Chris. -- http://www.bluishcoder.co.nz ---

Re: [Factor-talk] Git

2010-11-14 Thread Shaping
> git branch displays (no branch) master So Git GUI makes the master branch by default, but does not put you on it. Why is that a good thing to do? Shaping -- Centralized Desktop Delivery: Dell and VMware Reference A

Re: [Factor-talk] Git

2010-11-14 Thread Chris Double
On Mon, Nov 15, 2010 at 12:47 PM, Shaping wrote: > Some of the doc says that your default branch is "master".  This is not > correct.  I just did a "git status" and see "Not currently on a branch" What they mean is when you first clone a repository you get a default branch called 'master'. And th

Re: [Factor-talk] Git

2010-11-14 Thread Shaping
Some of the doc says that your default branch is "master". This is not correct. I just did a "git status" and see "Not currently on a branch" Shaping -Original Message- From: William Tanksley, Jr [mailto:wtanksle...@gmail.com] Sent: 2010-November-14, 17:42 To: factor-talk@lists.source

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Shaping
Yes. By "directory" below is meant place of the "repo", whether a local place or a remote one. Shaping -Original Message- From: William Tanksley, Jr [mailto:wtanksle...@gmail.com] Sent: 2010-November-14, 17:42 To: factor-talk@lists.sourceforge.net Subject: Re: [Factor-talk] Furnace on W

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread William Tanksley, Jr
Shaping wrote: >> If you have a personal git host somewhere, and there are free ones, try >> making two clones of the same little folder > I thought one clones only a repo. Every working directory cloned from a repo is itself a repo. This is why git is a "decentralized" and "distributed" version

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Shaping
You are managing interaction between three places, and I'm not great on using the best words, so listen for the themes as I explain my working metal model for git :) . A) The remote place, B) your local git version of that, and C) your physical file layout. Commands like fetch work between A and

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Shaping
There's a lot going on, so context is difficult here, but I was trying to go back in time before you made changes, and suggest when working on Factor main code, you make changes *in a branch,* sync "sync" means to resolve conflicting changes? I am reading the Git documentation from beginni

[Factor-talk] What should I call to issue [ 8080 httpd ] in-thread again, after closing factor & reopening?

2010-11-14 Thread Jim mack
reopening factor, then calling to [ 8080 httpd ] in-thread is now throwing this kind of an error [2010-11-14T19:53:33Z] NOTICE start-accept-loop: T{ inet6 f "::" 8080 } [2010-11-14T19:53:33Z] NOTICE start-accept-loop: T{ inet4 f "0.0.0.0" 8080 } [2010-11-14T19:53:33Z] ERROR start-accept-loop: Uni

Re: [Factor-talk] linode memory problem

2010-11-14 Thread Jim mack
I'm running ubuntu 10.4 lts, and just refreshed factor & rebuilt using the ./build-support/factor.sh update method at about 10pm Saturday. The last time I build was 2 months ago, and the code has changed, but runs on both pc & mac using the binary downloads as recent as a week ago. On Sun, Nov 14,

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Jim mack
On Sun, Nov 14, 2010 at 1:08 AM, Shaping wrote: > > So why does this thing called a "fetch" exist, if it does not change repo > state? > You are managing interaction between three places, and I'm not great on using the best words, so listen for the themes as I explain my working metal model for g

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Jim mack
There's a lot going on, so context is difficult here, but I was trying to go back in time before you made changes, and suggest when working on Factor main code, you make changes *in a branch,* sync with the larger community in the main trunc, not a branch, then integrate your stuff by merging from

[Factor-talk] Arrow models patch

2010-11-14 Thread Jon Harper
Hi everyone, what do you think about removing the "model" slot from arrow models? It looks redundant with the dependencies slot to me. You can take a look at the attached patch that removes it (unit-tests pass with this patch) Jon From 75a9483dc8df5f24ee94118349cf303531354375 Mon Sep 17 00:00:00 2

Re: [Factor-talk] Git

2010-11-14 Thread Shaping
Thanks. I'll do some reading. Building a mental model for how Git works from Git GUI is not working as well as I want. Shaping -Original Message- From: Chris Double [mailto:chris.dou...@double.co.nz] Sent: 2010-November-14, 04:41 To: factor-talk@lists.sourceforge.net Subject: Re: [F

Re: [Factor-talk] Git checking-out, updating, fetching; adding the work directory

2010-11-14 Thread Shaping
I have a .def file and a deploy.factor file remaining, uncommitted. I'm not sure what to do with these two. The .def file can always be re-sourced from the downloaded zip file. The deploy.factor is auto-generated during deployment of the Tetris demo. I don't see why I should commit it. I don't

Re: [Factor-talk] Git

2010-11-14 Thread Chris Double
On Sun, Nov 14, 2010 at 11:19 PM, Shaping wrote: > ...stores the code but not in your repo directories.  The code is retained > in some Git data file in the installation direction, I suppose. Yes. > > origin == original clone?  I guess not. 'origin' is the remote repository you originally clone

Re: [Factor-talk] Git

2010-11-14 Thread Shaping
On Sun, Nov 14, 2010 at 10:40 PM, Shaping wrote: > Are you saying that fetch downloads a model of some remote committed code, > so that I can somehow view an abbreviated representation of that code, so > that I can later, at my convenience, select some or all of it and then > actually download

Re: [Factor-talk] Git GUI commit: deploy.factor

2010-11-14 Thread Shaping
What is deploy.factor all about? It looks like a list of vocabs to load before building the exe. Do we usually commit stuff like this? Probably not. Shaping From: Shaping [mailto:shap...@charter.net] Sent: 2010-November-14, 03:44 To: factor-talk@lists.sourceforge.net Subject: Re: [Fa

Re: [Factor-talk] Git GUI commit

2010-11-14 Thread Shaping
...Okay you can Commit => Ammend Last Commit to fix mistakes like this... Shaping From: Shaping [mailto:shap...@charter.net] Sent: 2010-November-14, 03:44 To: factor-talk@lists.sourceforge.net Subject: Re: [Factor-talk] Git GUI commit Well, that was a little unexpected. All file

Re: [Factor-talk] Git GUI commit

2010-11-14 Thread Shaping
Well, that was a little unexpected. All files in the commit window are committed under a comment I wrote for just one of them. I had only one file selected. I guess you're expected to commit one at a time if you need to individuate comments, which I do. Now I need to roll-back the last commit..

Re: [Factor-talk] Git fetch

2010-11-14 Thread Chris Double
On Sun, Nov 14, 2010 at 10:40 PM, Shaping wrote: > Are you saying that fetch downloads a model of some remote committed code, > so that I can somehow view an abbreviated representation of that code, so > that I can later, at my convenience, select some or all of it and then > actually download the

Re: [Factor-talk] Git fetch

2010-11-14 Thread Shaping
> So why does this thing called a "fetch" exist, if it does not change repo > state? > It downloads all commits from the remote repository that you don't already have and stores it in a remote branch in your repository. What it doesn't do is change any of your existing commits or code. It

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Chris Double
On Sun, Nov 14, 2010 at 10:08 PM, Shaping wrote: > > So why does this thing called a "fetch" exist, if it does not change repo > state? > It downloads all commits from the remote repository that you don't already have and stores it in a remote branch in your repository. What it doesn't do is chan

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Shaping
> I know. I'm concerned about possible collisions with modified stock code. > I think the point Chris is making in his step 2 is that these changes need > to be committed, first, but he did not mention old directories , only new > ones. 'git add' is used to tell git about new files you have

Re: [Factor-talk] Git GUI help page

2010-11-14 Thread Shaping
http://nathanj.github.com/gitguide/tour.html This seems to be the only good explanation of Git GUI. Shaping -Original Message- From: Chris Double [mailto:chris.dou...@double.co.nz] Sent: 2010-November-14, 01:55 To: factor-talk@lists.sourceforge.net Subject: Re: [Factor-talk] Furnace

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Shaping
You don't need to recover from a 'fetch'. What that does is it downloads the stuff you don't yet have and stores it internally in a 'remote' branch. It makes no changes at all to your checked out code or your changes. That only happens when you do a 'rebase', 'merge' or 'pull'. pull == chec

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Chris Double
On Sun, Nov 14, 2010 at 10:01 PM, Shaping wrote: > I know.  I'm concerned about possible collisions with modified stock code. > I think the point Chris is making in his step 2 is that these changes need > to be committed, first, but he did not mention old directories , only new > ones. 'git add'

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Chris Double
On Sun, Nov 14, 2010 at 9:55 PM, Shaping wrote: > Right now I need to recover from a "fetch origin". You don't need to recover from a 'fetch'. What that does is it downloads the stuff you don't yet have and stores it internally in a 'remote' branch. It makes no changes at all to your checked out

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Shaping
the work repository is a separate git base from the larger code base. I know. This is the result of the cloning. When you are in factor\work, git knows about your personal repository (assuming you have one set up) but when you are in factor, it knows about where you cloned from, but will ignor

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Shaping
http://book.git-scm.com/3_basic_branching_and_merging.html you can also do 1 git branch playing 2 git checkout playing 3 -- perform changes 4 git checkout master -- you're back to level 2, but 3 is avail How is 3 available? Is it version separately from the master you just restored on

Re: [Factor-talk] Furnace

2010-11-14 Thread Shaping
Are they Win 32 or Win 64 versions? I'm not sure. Where did you get the files from? http://www.sqlite.org/download.html Do they have the same names as what the sqlite vocab is expecting? ffi.factor contains a reference to sqlite3.dll. "sqlite3

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Jim mack
the work repository is a separate git base from the larger code base. When you are in factor\work, git knows about your personal repository (assuming you have one set up) but when you are in factor, it knows about where you cloned from, but will ignore work as that's a personal distribution. On S

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Shaping
I only know the command line so I'll give you command line tips and you can translate them to equivalent GUI commands. 1) Clone the factor respository git clone git://factorcode.org/git/factor.git 2) In this reposi

Re: [Factor-talk] Furnace on Windows

2010-11-14 Thread Jim mack
http://book.git-scm.com/3_basic_branching_and_merging.html you can also do 1 git branch playing 2 git checkout playing 3 -- perform changes 4 git checkout master -- you're back to level 2, but 3 is avail 5 rebuild as above 6 git checkout playing -- you're back to your playing git is really cool a

[Factor-talk] linode memory problem

2010-11-14 Thread Jim mack
How do I go about starting to think about this: ( scratchpad - auto ) Out of memory Nursery: Start=b6c8, size=10, end=b6d8 Aging: Start=b6a8, size=20, end=b6c8 Tenured: Start=b088, size=600, end=b688 Cards:base=b080a008, size=65400 r...@blogtation:/usr/factor#