workspace stringclasses 1
value | channel stringclasses 1
value | sentences stringlengths 1 3.93k | ts stringlengths 26 26 | user stringlengths 2 11 | sentence_id stringlengths 44 53 | timestamp float64 1.5B 1.56B | __index_level_0__ int64 0 106k |
|---|---|---|---|---|---|---|---|
pythondev | help | but its still a log, so dont stress about it, you should at minimum track the who was paid what when stuff | 2017-09-15T15:56:39.000181 | Orpha | pythondev_help_Orpha_2017-09-15T15:56:39.000181 | 1,505,490,999.000181 | 93,903 |
pythondev | help | yea, you should keep your own log too though | 2017-09-15T15:56:55.000618 | Orpha | pythondev_help_Orpha_2017-09-15T15:56:55.000618 | 1,505,491,015.000618 | 93,904 |
pythondev | help | But the reason I ask is when customers change plans the easiest way for me to do it was to have them cancel their current subscription and sign up for a different one, rather than use the stripe upgrade/downgrade stuff and deal with changing billing cycles etc | 2017-09-15T15:57:25.000550 | Huey | pythondev_help_Huey_2017-09-15T15:57:25.000550 | 1,505,491,045.00055 | 93,905 |
pythondev | help | maybe your system things someone got paid, and stripe doesnt show a refund… this is why you also need a log | 2017-09-15T15:57:38.000127 | Orpha | pythondev_help_Orpha_2017-09-15T15:57:38.000127 | 1,505,491,058.000127 | 93,906 |
pythondev | help | so when I do that and delete the customer to start fresh it wipes their billing history. | 2017-09-15T15:57:39.000151 | Huey | pythondev_help_Huey_2017-09-15T15:57:39.000151 | 1,505,491,059.000151 | 93,907 |
pythondev | help | ok, cool | 2017-09-15T15:57:50.000133 | Huey | pythondev_help_Huey_2017-09-15T15:57:50.000133 | 1,505,491,070.000133 | 93,908 |
pythondev | help | I will probably create a payment history table for each user then | 2017-09-15T15:58:10.000312 | Huey | pythondev_help_Huey_2017-09-15T15:58:10.000312 | 1,505,491,090.000312 | 93,909 |
pythondev | help | ah, you should for sure use stripes (or your own) upgrade/downgrade | 2017-09-15T15:58:32.000043 | Orpha | pythondev_help_Orpha_2017-09-15T15:58:32.000043 | 1,505,491,112.000043 | 93,910 |
pythondev | help | dont let users cancel then signup again | 2017-09-15T15:58:41.000266 | Orpha | pythondev_help_Orpha_2017-09-15T15:58:41.000266 | 1,505,491,121.000266 | 93,911 |
pythondev | help | you will likely lose people that way | 2017-09-15T15:58:48.000310 | Orpha | pythondev_help_Orpha_2017-09-15T15:58:48.000310 | 1,505,491,128.00031 | 93,912 |
pythondev | help | well, then I will need to undo everything I just did, lol | 2017-09-15T15:59:18.000191 | Huey | pythondev_help_Huey_2017-09-15T15:59:18.000191 | 1,505,491,158.000191 | 93,913 |
pythondev | help | lol | 2017-09-15T15:59:21.000643 | Orpha | pythondev_help_Orpha_2017-09-15T15:59:21.000643 | 1,505,491,161.000643 | 93,914 |
pythondev | help | i mean, its your app.. so you can do what you want | 2017-09-15T15:59:35.000189 | Orpha | pythondev_help_Orpha_2017-09-15T15:59:35.000189 | 1,505,491,175.000189 | 93,915 |
pythondev | help | It takes about 2 seconds to switch though, I couldn't see people really having an issue with that | 2017-09-15T15:59:43.000050 | Huey | pythondev_help_Huey_2017-09-15T15:59:43.000050 | 1,505,491,183.00005 | 93,916 |
pythondev | help | just doesnt sound like the best UX | 2017-09-15T15:59:46.000137 | Orpha | pythondev_help_Orpha_2017-09-15T15:59:46.000137 | 1,505,491,186.000137 | 93,917 |
pythondev | help | im no expert though | 2017-09-15T15:59:59.000095 | Orpha | pythondev_help_Orpha_2017-09-15T15:59:59.000095 | 1,505,491,199.000095 | 93,918 |
pythondev | help | so yea haha | 2017-09-15T16:00:02.000481 | Orpha | pythondev_help_Orpha_2017-09-15T16:00:02.000481 | 1,505,491,202.000481 | 93,919 |
pythondev | help | do you have a subscriptions table? | 2017-09-15T16:00:39.000062 | Orpha | pythondev_help_Orpha_2017-09-15T16:00:39.000062 | 1,505,491,239.000062 | 93,920 |
pythondev | help | could you not just ‘on upgrade’ switch the users subscription_id to whatever they changed to | 2017-09-15T16:01:31.000258 | Orpha | pythondev_help_Orpha_2017-09-15T16:01:31.000258 | 1,505,491,291.000258 | 93,921 |
pythondev | help | then in transactions log you can also store what subscription they were on at the time of the transaction | 2017-09-15T16:01:54.000123 | Orpha | pythondev_help_Orpha_2017-09-15T16:01:54.000123 | 1,505,491,314.000123 | 93,922 |
pythondev | help | Yeah I could do that. I might ... this was just the fastest/easiest way I could think to get it working | 2017-09-15T16:02:19.000605 | Huey | pythondev_help_Huey_2017-09-15T16:02:19.000605 | 1,505,491,339.000605 | 93,923 |
pythondev | help | for sure | 2017-09-15T16:02:30.000400 | Orpha | pythondev_help_Orpha_2017-09-15T16:02:30.000400 | 1,505,491,350.0004 | 93,924 |
pythondev | help | hey guys, is this a good way to sum properties of objects contained in two different arrays?
```
expenses = [Expense("pizza", 50), Expense("Rock Concert", 100)]
incomes = [Income("Salary", 5000), Income("Dividends", 200)]
total = (
reduce(add, map(lambda income: income.amount, incomes)) -
reduce(add, map(lambda expense: expense.amount, expenses))
)
``` | 2017-09-15T21:17:55.000014 | Latarsha | pythondev_help_Latarsha_2017-09-15T21:17:55.000014 | 1,505,510,275.000014 | 93,925 |
pythondev | help | ```sum(income.amount for income in incomes) - sum(expense.amount for expense in expenses)``` ? | 2017-09-16T04:45:17.000064 | Cristy | pythondev_help_Cristy_2017-09-16T04:45:17.000064 | 1,505,537,117.000064 | 93,926 |
pythondev | help | traditionally, this is done by having one list of transactions, and the expenses having a negative value, this way you can just sum one list, it also makes several other common operations much simpler | 2017-09-16T04:47:29.000065 | Cristy | pythondev_help_Cristy_2017-09-16T04:47:29.000065 | 1,505,537,249.000065 | 93,927 |
pythondev | help | very nice, thanks <@Cristy> | 2017-09-16T08:57:33.000012 | Latarsha | pythondev_help_Latarsha_2017-09-16T08:57:33.000012 | 1,505,552,253.000012 | 93,928 |
pythondev | help | Hello, someone knows how to iterate in Numpy.matrix ? like:
x = np.matrix([1,2,3]) | 2017-09-16T11:56:40.000021 | Qiana | pythondev_help_Qiana_2017-09-16T11:56:40.000021 | 1,505,563,000.000021 | 93,929 |
pythondev | help | <@Qiana> is it just `for item in x.flat`? | 2017-09-16T13:30:04.000001 | Mallie | pythondev_help_Mallie_2017-09-16T13:30:04.000001 | 1,505,568,604.000001 | 93,930 |
pythondev | help | <https://docs.scipy.org/doc/numpy-1.13.0/reference/generated/numpy.matrix.html> | 2017-09-16T13:30:05.000127 | Mallie | pythondev_help_Mallie_2017-09-16T13:30:05.000127 | 1,505,568,605.000127 | 93,931 |
pythondev | help | oh thanks, <@Mallie> i figured out how to work with it. I'm starting to work with numpy :weary: | 2017-09-16T15:05:09.000051 | Qiana | pythondev_help_Qiana_2017-09-16T15:05:09.000051 | 1,505,574,309.000051 | 93,932 |
pythondev | help | Got a question about `threading.Thread` (_I'm fairly new to working with threads & am not 100% sure about what's going on behind the scenes_) .
I'm generating some live MIDI signals, and found that if I use `thread.start(); thread.join();`, the thread becomes blocking and I lose realtime functionality.
Alternatively, if I only use `thead.start()`, without `thread.join()`, the realtime functionality acts as expected.
Are there any potential issues with starting a thread and never joining it?
Is joining an optional thing for when you _want_ a thread to be blocking, or is its usage required in some way? | 2017-09-16T22:27:23.000008 | Libby | pythondev_help_Libby_2017-09-16T22:27:23.000008 | 1,505,600,843.000008 | 93,933 |
pythondev | help | <@Libby> I have not used the `threading` module much but I can't see anything in the doc to imply that `thread.join()` is is anyway a necessary operation. The thread terminates when `run()` returns, you shouldn't have to do more than ensure there isn't an infinite loop somewhere in there to "clean up" the thread part of things. You definitely shouldn't _have_ to force another thread to wait for it with `join()`. | 2017-09-16T22:43:50.000039 | Mallie | pythondev_help_Mallie_2017-09-16T22:43:50.000039 | 1,505,601,830.000039 | 93,934 |
pythondev | help | I have a maybe dumb question about the sortedcontainers package. is there a performance advantage to using sortedlist versus a regular python list and sorting after each insertion? as i understand it, sorting is a solved problem at O(nlogn) so whatever sortedlist does to maintain sorting at all times cant be any faster then the standardlib sorting function, right? | 2017-09-17T03:15:59.000034 | Catheryn | pythondev_help_Catheryn_2017-09-17T03:15:59.000034 | 1,505,618,159.000034 | 93,935 |
pythondev | help | It can be. | 2017-09-17T04:20:01.000037 | Suellen | pythondev_help_Suellen_2017-09-17T04:20:01.000037 | 1,505,622,001.000037 | 93,936 |
pythondev | help | If you insert and then sort, you're doing `n * n * logn` which is `n^2` | 2017-09-17T04:20:21.000011 | Suellen | pythondev_help_Suellen_2017-09-17T04:20:21.000011 | 1,505,622,021.000011 | 93,937 |
pythondev | help | if you insert something into already sorted container, you kind of already know which position it will occupy (maybe not with 100% certainty) | 2017-09-17T04:21:06.000034 | Suellen | pythondev_help_Suellen_2017-09-17T04:21:06.000034 | 1,505,622,066.000034 | 93,938 |
pythondev | help | if you just append to the list then sort its 1 + n * logn, right? | 2017-09-17T04:36:05.000046 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:36:05.000046 | 1,505,622,965.000046 | 93,939 |
pythondev | help | whats the performance of binary search using bisection? i think thats what sortedcontainers library uses | 2017-09-17T04:38:48.000054 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:38:48.000054 | 1,505,623,128.000054 | 93,940 |
pythondev | help | if you append every time it's N append+sort operations, and with sort itself being n*logn it becomes n^2 | 2017-09-17T04:40:00.000008 | Suellen | pythondev_help_Suellen_2017-09-17T04:40:00.000008 | 1,505,623,200.000008 | 93,941 |
pythondev | help | oh right | 2017-09-17T04:40:28.000029 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:40:28.000029 | 1,505,623,228.000029 | 93,942 |
pythondev | help | bisect is logn so, if you use it to maintain the sort then its n * logn, right? | 2017-09-17T04:41:24.000038 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:41:24.000038 | 1,505,623,284.000038 | 93,943 |
pythondev | help | :thinking_face: | 2017-09-17T04:44:04.000015 | Suellen | pythondev_help_Suellen_2017-09-17T04:44:04.000015 | 1,505,623,444.000015 | 93,944 |
pythondev | help | ```bisect.insort_left(a, x, lo=0, hi=len(a))¶
Insert x in a in sorted order. This is equivalent to a.insert(bisect.bisect_left(a, x, lo, hi), x) assuming that a is already sorted. Keep in mind that the O(log n) search is dominated by the slow O(n) insertion step.``` | 2017-09-17T04:44:18.000041 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:44:18.000041 | 1,505,623,458.000041 | 93,945 |
pythondev | help | Not sure if this library does this, but one trick about sorted containers is that they don't actually need to be sorted at all times or after every insert. They only have to be sorted when you try to read from them | 2017-09-17T04:45:42.000009 | Gabriele | pythondev_help_Gabriele_2017-09-17T04:45:42.000009 | 1,505,623,542.000009 | 93,946 |
pythondev | help | So they can postpone the sorts, unless they're trying to provide strong guarantees about each operation | 2017-09-17T04:46:08.000046 | Gabriele | pythondev_help_Gabriele_2017-09-17T04:46:08.000046 | 1,505,623,568.000046 | 93,947 |
pythondev | help | thats a good point | 2017-09-17T04:46:42.000004 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:46:42.000004 | 1,505,623,602.000004 | 93,948 |
pythondev | help | but the SortedContainers package maintains the order at all times | 2017-09-17T04:46:56.000017 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:46:56.000017 | 1,505,623,616.000017 | 93,949 |
pythondev | help | <http://www.grantjenks.com/docs/sortedcontainers/introduction.html> | 2017-09-17T04:46:57.000017 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:46:57.000017 | 1,505,623,617.000017 | 93,950 |
pythondev | help | best just benchmark it | 2017-09-17T04:47:07.000022 | Suellen | pythondev_help_Suellen_2017-09-17T04:47:07.000022 | 1,505,623,627.000022 | 93,951 |
pythondev | help | good call. i’ll do that now | 2017-09-17T04:49:42.000032 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:49:42.000032 | 1,505,623,782.000032 | 93,952 |
pythondev | help | It looks like it uses something like a heap of indices. Personally I'd be surprised if this sort of package made a positive difference to performance, with the extra memory and complexity involved. | 2017-09-17T04:50:10.000018 | Gabriele | pythondev_help_Gabriele_2017-09-17T04:50:10.000018 | 1,505,623,810.000018 | 93,953 |
pythondev | help | i saw it mentioned on a lot of stackoverflow questions about sorted datastructures so i figured it was considered the standard | 2017-09-17T04:51:27.000018 | Catheryn | pythondev_help_Catheryn_2017-09-17T04:51:27.000018 | 1,505,623,887.000018 | 93,954 |
pythondev | help | probably is. :slightly_smiling_face: | 2017-09-17T04:51:46.000035 | Gabriele | pythondev_help_Gabriele_2017-09-17T04:51:46.000035 | 1,505,623,906.000035 | 93,955 |
pythondev | help | okay sorted() is blazingly fast | 2017-09-17T05:01:31.000020 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:01:31.000020 | 1,505,624,491.00002 | 93,956 |
pythondev | help | oh wait | 2017-09-17T05:02:31.000044 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:02:31.000044 | 1,505,624,551.000044 | 93,957 |
pythondev | help | derp, i didnt sort between each insert in normalListInsert | 2017-09-17T05:02:40.000008 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:02:40.000008 | 1,505,624,560.000008 | 93,958 |
pythondev | help | only did one insert at theend | 2017-09-17T05:02:43.000057 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:02:43.000057 | 1,505,624,563.000057 | 93,959 |
pythondev | help | gotta rerun it | 2017-09-17T05:04:58.000005 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:04:58.000005 | 1,505,624,698.000005 | 93,960 |
pythondev | help | sorting() after each insert is terrible actually | 2017-09-17T05:05:34.000013 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:05:34.000013 | 1,505,624,734.000013 | 93,961 |
pythondev | help | ```import random
import bisect
from sortedcontainers import SortedList
from pyutils import timeit
@timeit
def normalListInplace(arr: list) -> list:
return sorted(arr)
@timeit
def normalListInsert(arr: list) -> list:
result = []
for el in arr:
result.append(el)
result = sorted(result)
return result
@timeit
def bisectListInsert(arr: list) -> list:
result = []
for el in arr:
bisect.insort_left(result, el, lo=0, hi=len(result))
return result
@timeit
def sortedListInsert(arr: list) -> list:
result = SortedList()
for el in arr:
result.add(el)
return result
if __name__== '__main__':
L = random.sample(range(0,10000), 10000)
normalListInplace(L)
normalListInsert(L)
bisectListInsert(L)
sortedListInsert(L)
``` | 2017-09-17T05:05:39.000028 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:05:39.000028 | 1,505,624,739.000028 | 93,962 |
pythondev | help | > sorting() after each insert is terrible actually
no way :stuck_out_tongue: | 2017-09-17T05:05:50.000039 | Suellen | pythondev_help_Suellen_2017-09-17T05:05:50.000039 | 1,505,624,750.000039 | 93,963 |
pythondev | help | ```solomon@MacBookPro python 💰 python list_sorting.py
'normalListInplace' 2.90 ms
'normalListInsert' 1232.83 ms
'bisectListInsert' 33.03 ms
'sortedListInsert' 33.08 ms
``` | 2017-09-17T05:05:51.000011 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:05:51.000011 | 1,505,624,751.000011 | 93,964 |
pythondev | help | couldnt even run it with 100,000 elements in the list | 2017-09-17T05:06:22.000035 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:06:22.000035 | 1,505,624,782.000035 | 93,965 |
pythondev | help | bisect matches sortedList package up to 10,000 elements then slows down substantially | 2017-09-17T05:08:20.000036 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:08:20.000036 | 1,505,624,900.000036 | 93,966 |
pythondev | help | at 100,000 elements its 1982ms vs 348ms | 2017-09-17T05:08:38.000004 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:08:38.000004 | 1,505,624,918.000004 | 93,967 |
pythondev | help | now i need to read the source of sortedList to see what they are doing | 2017-09-17T05:09:10.000005 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:09:10.000005 | 1,505,624,950.000005 | 93,968 |
pythondev | help | oh neat, it plays on the fact that bisect.insort is fast up to 10,000 elements | 2017-09-17T05:14:59.000020 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:14:59.000020 | 1,505,625,299.00002 | 93,969 |
pythondev | help | and splits the list into multiple sublists that are load balanced | 2017-09-17T05:15:17.000014 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:15:17.000014 | 1,505,625,317.000014 | 93,970 |
pythondev | help | with a bunch more magic on top of that | 2017-09-17T05:16:55.000043 | Catheryn | pythondev_help_Catheryn_2017-09-17T05:16:55.000043 | 1,505,625,415.000043 | 93,971 |
pythondev | help | Out of interest, might heapq be a contender here? | 2017-09-17T06:52:28.000013 | Cristy | pythondev_help_Cristy_2017-09-17T06:52:28.000013 | 1,505,631,148.000013 | 93,972 |
pythondev | help | <@Catheryn> It makes sense that the sortedList is as fast as bisect, as it uses it internally | 2017-09-17T07:58:28.000023 | Tatum | pythondev_help_Tatum_2017-09-17T07:58:28.000023 | 1,505,635,108.000023 | 93,973 |
pythondev | help | <https://github.com/grantjenks/sorted_containers/blob/master/sortedcontainers/sortedlist.py#L8> | 2017-09-17T07:58:30.000022 | Tatum | pythondev_help_Tatum_2017-09-17T07:58:30.000022 | 1,505,635,110.000022 | 93,974 |
pythondev | help | I honestly wonder why none of these are implemented using a C-written balanced binary search tree | 2017-09-17T08:00:12.000001 | Tatum | pythondev_help_Tatum_2017-09-17T08:00:12.000001 | 1,505,635,212.000001 | 93,975 |
pythondev | help | <@Tatum> would that be substantially faster? | 2017-09-17T12:32:04.000086 | Suellen | pythondev_help_Suellen_2017-09-17T12:32:04.000086 | 1,505,651,524.000086 | 93,976 |
pythondev | help | I can’t tell because I’m not sure of the running time on the bisect algorithm | 2017-09-17T12:46:03.000021 | Tatum | pythondev_help_Tatum_2017-09-17T12:46:03.000021 | 1,505,652,363.000021 | 93,977 |
pythondev | help | but using a tree would be O(log n) per element addition | 2017-09-17T12:46:16.000073 | Tatum | pythondev_help_Tatum_2017-09-17T12:46:16.000073 | 1,505,652,376.000073 | 93,978 |
pythondev | help | which is super fast | 2017-09-17T12:46:21.000094 | Tatum | pythondev_help_Tatum_2017-09-17T12:46:21.000094 | 1,505,652,381.000094 | 93,979 |
pythondev | help | <@Tatum> well sortedlist is marketed as a pure Python solution | 2017-09-17T13:57:20.000032 | Catheryn | pythondev_help_Catheryn_2017-09-17T13:57:20.000032 | 1,505,656,640.000032 | 93,980 |
pythondev | help | a pure python tree should be faster | 2017-09-17T13:57:53.000097 | Tatum | pythondev_help_Tatum_2017-09-17T13:57:53.000097 | 1,505,656,673.000097 | 93,981 |
pythondev | help | Hmm. I have an avl tree I wrote for a project. I'll compare that | 2017-09-17T14:00:26.000033 | Catheryn | pythondev_help_Catheryn_2017-09-17T14:00:26.000033 | 1,505,656,826.000033 | 93,982 |
pythondev | help | <https://github.com/ssbothwell/BinPack/blob/master/binpack/avl_tree.py> | 2017-09-17T14:01:11.000057 | Catheryn | pythondev_help_Catheryn_2017-09-17T14:01:11.000057 | 1,505,656,871.000057 | 93,983 |
pythondev | help | It's fast but it depends on how much rebalancing ends up happening | 2017-09-17T14:01:33.000093 | Catheryn | pythondev_help_Catheryn_2017-09-17T14:01:33.000093 | 1,505,656,893.000093 | 93,984 |
pythondev | help | I wish the standard library had a bunch of trees built in | 2017-09-17T14:03:24.000049 | Catheryn | pythondev_help_Catheryn_2017-09-17T14:03:24.000049 | 1,505,657,004.000049 | 93,985 |
pythondev | help | <@Cristy> heapq could be a contender if
You only need access to the min or max value of the list. For fun I'll timeit as well | 2017-09-17T14:04:20.000109 | Catheryn | pythondev_help_Catheryn_2017-09-17T14:04:20.000109 | 1,505,657,060.000109 | 93,986 |
pythondev | help | Actually no | 2017-09-17T14:12:44.000074 | Tatum | pythondev_help_Tatum_2017-09-17T14:12:44.000074 | 1,505,657,564.000074 | 93,987 |
pythondev | help | a tree in python turns out to be slower | 2017-09-17T14:12:59.000102 | Tatum | pythondev_help_Tatum_2017-09-17T14:12:59.000102 | 1,505,657,579.000102 | 93,988 |
pythondev | help | I think this is due to the class instantiation overhead | 2017-09-17T14:13:06.000062 | Tatum | pythondev_help_Tatum_2017-09-17T14:13:06.000062 | 1,505,657,586.000062 | 93,989 |
pythondev | help | ```
RB Tree took 0:00:01.798895
Sorted list took 0:00:00.258929
```
with 100k elements | 2017-09-17T14:13:20.000048 | Tatum | pythondev_help_Tatum_2017-09-17T14:13:20.000048 | 1,505,657,600.000048 | 93,990 |
pythondev | help | Yah try using. _slot_ to remove the objects dict | 2017-09-17T14:15:03.000083 | Catheryn | pythondev_help_Catheryn_2017-09-17T14:15:03.000083 | 1,505,657,703.000083 | 93,991 |
pythondev | help | I don’t think its a memory problem | 2017-09-17T14:15:12.000064 | Tatum | pythondev_help_Tatum_2017-09-17T14:15:12.000064 | 1,505,657,712.000064 | 93,992 |
pythondev | help | but this makes me think that bisect is O(log N) as well | 2017-09-17T14:15:41.000052 | Tatum | pythondev_help_Tatum_2017-09-17T14:15:41.000052 | 1,505,657,741.000052 | 93,993 |
pythondev | help | <https://stackoverflow.com/a/12022278/3218627> | 2017-09-17T14:16:22.000046 | Tatum | pythondev_help_Tatum_2017-09-17T14:16:22.000046 | 1,505,657,782.000046 | 93,994 |
pythondev | help | Ah there we go, yeah no wonder its faster | 2017-09-17T14:16:29.000052 | Tatum | pythondev_help_Tatum_2017-09-17T14:16:29.000052 | 1,505,657,789.000052 | 93,995 |
pythondev | help | Significant overhead by the tree class | 2017-09-17T14:16:47.000010 | Tatum | pythondev_help_Tatum_2017-09-17T14:16:47.000010 | 1,505,657,807.00001 | 93,996 |
pythondev | help | wait this isnt fair | 2017-09-17T14:21:14.000016 | Tatum | pythondev_help_Tatum_2017-09-17T14:21:14.000016 | 1,505,658,074.000016 | 93,997 |
pythondev | help | the `bisect` module which is the core of `SortedList` is implemented in C as well | 2017-09-17T14:21:30.000090 | Tatum | pythondev_help_Tatum_2017-09-17T14:21:30.000090 | 1,505,658,090.00009 | 93,998 |
pythondev | help | <https://github.com/python/cpython/blob/3.6/Lib/bisect.py#L90> | 2017-09-17T14:21:46.000088 | Tatum | pythondev_help_Tatum_2017-09-17T14:21:46.000088 | 1,505,658,106.000088 | 93,999 |
pythondev | help | I can’t find it on their github but I have the library executable on my local machine
`/Library/Frameworks/Python.framework/Versions/3.6/lib/python3.6/lib-dynload/_bisect.cpython-36m-darwin.so`
I found this by opening the `_bisect` file through Pycharm | 2017-09-17T14:23:06.000003 | Tatum | pythondev_help_Tatum_2017-09-17T14:23:06.000003 | 1,505,658,186.000003 | 94,000 |
pythondev | help | well sortedcontainers is substantially faster then bisect past 10,000 items | 2017-09-17T14:25:03.000047 | Catheryn | pythondev_help_Catheryn_2017-09-17T14:25:03.000047 | 1,505,658,303.000047 | 94,001 |
pythondev | help | ```'normalListInplace' 6.81 ms
'bisectListInsert' 100.45 ms
'sortedListInsert' 65.39 ms
'avlInsert' 602.80 ms
'heapInsert' 4.89 ms``` | 2017-09-17T14:32:39.000032 | Catheryn | pythondev_help_Catheryn_2017-09-17T14:32:39.000032 | 1,505,658,759.000032 | 94,002 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.