@@ -206,40 +206,138 @@ leave out at most one of `A` and `B`, in which case it defaults to `HEAD`.
206
206
207
207
Detached HEAD
208
208
-------------
209
+ HEAD normally refers to a named branch (e.g. 'master'). Meanwhile, each
210
+ branch refers to a specific commit. Let's look at a repo with three
211
+ commits, one of them tagged, and with branch 'master' checked out:
209
212
210
- It is sometimes useful to be able to 'checkout' a commit that is
211
- not at the tip of one of your branches. The most obvious
212
- example is to check out the commit at a tagged official release
213
- point, like this:
213
+ ------------
214
+ HEAD (refers to branch 'master')
215
+ |
216
+ v
217
+ a---b---c branch 'master' (refers to commit 'c')
218
+ ^
219
+ |
220
+ tag 'v2.0' (refers to commit 'b')
221
+ ------------
222
+
223
+ When a commit is created in this state, the branch is updated to refer to
224
+ the new commit. Specifically, 'git commit' creates a new commit 'd', whose
225
+ parent is commit 'c', and then updates branch 'master' to refer to new
226
+ commit 'd'. HEAD still refers to branch 'master' and so indirectly now refers
227
+ to commit 'd':
214
228
215
229
------------
216
- $ git checkout v2.6.18
230
+ $ edit; git add; git commit
231
+
232
+ HEAD (refers to branch 'master')
233
+ |
234
+ v
235
+ a---b---c---d branch 'master' (refers to commit 'd')
236
+ ^
237
+ |
238
+ tag 'v2.0' (refers to commit 'b')
217
239
------------
218
240
219
- Earlier versions of git did not allow this and asked you to
220
- create a temporary branch using the `-b` option, but starting from
221
- version 1.5.0, the above command 'detaches' your HEAD from the
222
- current branch and directly points at the commit named by the tag
223
- (`v2.6.18` in the example above).
241
+ It is sometimes useful to be able to checkout a commit that is not at
242
+ the tip of any named branch, or even to create a new commit that is not
243
+ referenced by a named branch. Let's look at what happens when we
244
+ checkout commit 'b' (here we show two ways this may be done):
224
245
225
- You can use all git commands while in this state. You can use
226
- `git reset --hard $othercommit` to further move around, for
227
- example. You can make changes and create a new commit on top of
228
- a detached HEAD. You can even create a merge by using `git
229
- merge $othercommit`.
246
+ ------------
247
+ $ git checkout v2.0 # or
248
+ $ git checkout master^^
249
+
250
+ HEAD (refers to commit 'b')
251
+ |
252
+ v
253
+ a---b---c---d branch 'master' (refers to commit 'd')
254
+ ^
255
+ |
256
+ tag 'v2.0' (refers to commit 'b')
257
+ ------------
230
258
231
- The state you are in while your HEAD is detached is not recorded
232
- by any branch (which is natural --- you are not on any branch).
233
- What this means is that you can discard your temporary commits
234
- and merges by switching back to an existing branch (e.g. `git
235
- checkout master`), and a later `git prune` or `git gc` would
236
- garbage-collect them. If you did this by mistake, you can ask
237
- the reflog for HEAD where you were, e.g.
259
+ Notice that regardless of which checkout command we use, HEAD now refers
260
+ directly to commit 'b'. This is known as being in detached HEAD state.
261
+ It means simply that HEAD refers to a specific commit, as opposed to
262
+ referring to a named branch. Let's see what happens when we create a commit:
238
263
239
264
------------
240
- $ git log -g -2 HEAD
265
+ $ edit; git add; git commit
266
+
267
+ HEAD (refers to commit 'e')
268
+ |
269
+ v
270
+ e
271
+ /
272
+ a---b---c---d branch 'master' (refers to commit 'd')
273
+ ^
274
+ |
275
+ tag 'v2.0' (refers to commit 'b')
241
276
------------
242
277
278
+ There is now a new commit 'e', but it is referenced only by HEAD. We can
279
+ of course add yet another commit in this state:
280
+
281
+ ------------
282
+ $ edit; git add; git commit
283
+
284
+ HEAD (refers to commit 'f')
285
+ |
286
+ v
287
+ e---f
288
+ /
289
+ a---b---c---d branch 'master' (refers to commit 'd')
290
+ ^
291
+ |
292
+ tag 'v2.0' (refers to commit 'b')
293
+ ------------
294
+
295
+ In fact, we can perform all the normal git operations. But, let's look
296
+ at what happens when we then checkout master:
297
+
298
+ ------------
299
+ $ git checkout master
300
+
301
+ HEAD (refers to branch 'master')
302
+ e---f |
303
+ / v
304
+ a---b---c---d branch 'master' (refers to commit 'd')
305
+ ^
306
+ |
307
+ tag 'v2.0' (refers to commit 'b')
308
+ ------------
309
+
310
+ It is important to realize that at this point nothing refers to commit
311
+ 'f'. Eventually commit 'f' (and by extension commit 'e') will be deleted
312
+ by the routine git garbage collection process, unless we create a reference
313
+ before that happens. If we have not yet moved away from commit 'f',
314
+ any of these will create a reference to it:
315
+
316
+ ------------
317
+ $ git checkout -b foo <1>
318
+ $ git branch foo <2>
319
+ $ git tag foo <3>
320
+ ------------
321
+
322
+ <1> creates a new branch 'foo', which refers to commit 'f', and then
323
+ updates HEAD to refer to branch 'foo'. In other words, we'll no longer
324
+ be in detached HEAD state after this command.
325
+
326
+ <2> similarly creates a new branch 'foo', which refers to commit 'f',
327
+ but leaves HEAD detached.
328
+
329
+ <3> creates a new tag 'foo', which refers to commit 'f',
330
+ leaving HEAD detached.
331
+
332
+ If we have moved away from commit 'f', then we must first recover its object
333
+ name (typically by using git reflog), and then we can create a reference to
334
+ it. For example, to see the last two commits to which HEAD referred, we
335
+ can use either of these commands:
336
+
337
+ ------------
338
+ $ git reflog -2 HEAD # or
339
+ $ git log -g -2 HEAD
340
+ ------------
243
341
244
342
EXAMPLES
245
343
--------
0 commit comments