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 | It's using ansible, so slightly obfuscated, but should give a good example if you can read through that | 2017-07-20T00:40:36.973160 | Beula | pythondev_help_Beula_2017-07-20T00:40:36.973160 | 1,500,511,236.97316 | 86,403 |
pythondev | help | <@Alvin> Unfortunately after changing `from . import config` in `__init__.py` to `from broadway import config` I still receive the error:
```from broadway import create_app, models, util
ModuleNotFoundError: No module named 'broadway'```
Thanks anyway though. It was worth a shot. :slightly_smiling_face: | 2017-07-20T03:44:41.361673 | Thomasina | pythondev_help_Thomasina_2017-07-20T03:44:41.361673 | 1,500,522,281.361673 | 86,404 |
pythondev | help | `latitude = db.Column(db.Float, nullable=False, asdecimal=True)` gives me `Unknown arguements passed to Clumn: ['asdecimal']`...am I doing something wrong here? | 2017-07-20T03:50:00.463374 | Thomasina | pythondev_help_Thomasina_2017-07-20T03:50:00.463374 | 1,500,522,600.463374 | 86,405 |
pythondev | help | not sure how to pass an argument to `db.Float` when constructing a Column with ORM | 2017-07-20T03:52:49.517566 | Suellen | pythondev_help_Suellen_2017-07-20T03:52:49.517566 | 1,500,522,769.517566 | 86,406 |
pythondev | help | but you can achieve the same with: | 2017-07-20T03:52:58.520319 | Suellen | pythondev_help_Suellen_2017-07-20T03:52:58.520319 | 1,500,522,778.520319 | 86,407 |
pythondev | help | `latitute = db.Column(db.Decimal, nullable=False)` | 2017-07-20T03:53:11.524572 | Suellen | pythondev_help_Suellen_2017-07-20T03:53:11.524572 | 1,500,522,791.524572 | 86,408 |
pythondev | help | `db.Decimal` (or `db.Numeric`, which is the same) is basically `db.Float` with `asdecimal=True` | 2017-07-20T03:53:44.534928 | Suellen | pythondev_help_Suellen_2017-07-20T03:53:44.534928 | 1,500,522,824.534928 | 86,409 |
pythondev | help | <@Suellen> `db.Decimal` didn't work but `db.Numeric` did. Thanks! :smile: | 2017-07-20T05:17:58.416420 | Thomasina | pythondev_help_Thomasina_2017-07-20T05:17:58.416420 | 1,500,527,878.41642 | 86,410 |
pythondev | help | :slightly_smiling_face: | 2017-07-20T05:18:06.419171 | Suellen | pythondev_help_Suellen_2017-07-20T05:18:06.419171 | 1,500,527,886.419171 | 86,411 |
pythondev | help | Is it better to have a local Postgres database for local testing of my app, or just have Postgres connect to the remote database on the staging server? :thinking_face: | 2017-07-20T05:43:52.998947 | Thomasina | pythondev_help_Thomasina_2017-07-20T05:43:52.998947 | 1,500,529,432.998947 | 86,412 |
pythondev | help | I can think of pros and cons to both, but I think that it's likely one of them is considered best practice... | 2017-07-20T05:44:18.008500 | Thomasina | pythondev_help_Thomasina_2017-07-20T05:44:18.008500 | 1,500,529,458.0085 | 86,413 |
pythondev | help | It doesn't matter, although you want to be accessing a separate _database_ on it (i.e. `CREATE DATABASE testing`) | 2017-07-20T05:49:30.124075 | Gabriele | pythondev_help_Gabriele_2017-07-20T05:49:30.124075 | 1,500,529,770.124075 | 86,414 |
pythondev | help | <@Gabriele> A different database on the staging server though? Wouldn't that be redundant or am I missing something? | 2017-07-20T05:52:13.184402 | Thomasina | pythondev_help_Thomasina_2017-07-20T05:52:13.184402 | 1,500,529,933.184402 | 86,415 |
pythondev | help | I would say that it is good practice to have local, staging and production environments | 2017-07-20T05:52:30.190919 | Meg | pythondev_help_Meg_2017-07-20T05:52:30.190919 | 1,500,529,950.190919 | 86,416 |
pythondev | help | with everything as close to 1:1 as possible | 2017-07-20T05:52:36.193168 | Meg | pythondev_help_Meg_2017-07-20T05:52:36.193168 | 1,500,529,956.193168 | 86,417 |
pythondev | help | and not doing so is a major red flag | 2017-07-20T05:52:52.199066 | Meg | pythondev_help_Meg_2017-07-20T05:52:52.199066 | 1,500,529,972.199066 | 86,418 |
pythondev | help | You don't want your testing environment accidentally trashing your live or staging environment. | 2017-07-20T05:53:04.203613 | Gabriele | pythondev_help_Gabriele_2017-07-20T05:53:04.203613 | 1,500,529,984.203613 | 86,419 |
pythondev | help | seen that happen :slightly_smiling_face: | 2017-07-20T05:53:14.206993 | Meg | pythondev_help_Meg_2017-07-20T05:53:14.206993 | 1,500,529,994.206993 | 86,420 |
pythondev | help | <@Meg> So, as in having a separate database locally, on the staging dyno, and on the production dyno? (Already have the latter two) | 2017-07-20T05:53:19.209075 | Thomasina | pythondev_help_Thomasina_2017-07-20T05:53:19.209075 | 1,500,529,999.209075 | 86,421 |
pythondev | help | correct | 2017-07-20T05:53:24.210903 | Meg | pythondev_help_Meg_2017-07-20T05:53:24.210903 | 1,500,530,004.210903 | 86,422 |
pythondev | help | so, you work on a laptop, right? | 2017-07-20T05:53:36.215406 | Meg | pythondev_help_Meg_2017-07-20T05:53:36.215406 | 1,500,530,016.215406 | 86,423 |
pythondev | help | I work on my desktop at home, but switch to a laptop when I work remotely sometimes. | 2017-07-20T05:53:53.221285 | Thomasina | pythondev_help_Thomasina_2017-07-20T05:53:53.221285 | 1,500,530,033.221285 | 86,424 |
pythondev | help | what happens if you want to do a little bit of testing in an area with spotty wifi, eg a coffee shop with a saturated network | 2017-07-20T05:54:07.226380 | Meg | pythondev_help_Meg_2017-07-20T05:54:07.226380 | 1,500,530,047.22638 | 86,425 |
pythondev | help | if you have a local db, that's one less network resource you need to worry about | 2017-07-20T05:54:19.231219 | Meg | pythondev_help_Meg_2017-07-20T05:54:19.231219 | 1,500,530,059.231219 | 86,426 |
pythondev | help | That's a good point. Actually I sometimes work places without any internet. | 2017-07-20T05:54:46.241103 | Thomasina | pythondev_help_Thomasina_2017-07-20T05:54:46.241103 | 1,500,530,086.241103 | 86,427 |
pythondev | help | Guess that's been decided then. Thanks :smile: | 2017-07-20T05:54:59.246097 | Thomasina | pythondev_help_Thomasina_2017-07-20T05:54:59.246097 | 1,500,530,099.246097 | 86,428 |
pythondev | help | so you can keep a small portion of the prod db on your local machine | 2017-07-20T05:55:06.248747 | Meg | pythondev_help_Meg_2017-07-20T05:55:06.248747 | 1,500,530,106.248747 | 86,429 |
pythondev | help | don't need to replicate the entire thing fully, especially if you have multi-gigs of records | 2017-07-20T05:55:28.256913 | Meg | pythondev_help_Meg_2017-07-20T05:55:28.256913 | 1,500,530,128.256913 | 86,430 |
pythondev | help | just enough to have testing and data integrity :slightly_smiling_face: | 2017-07-20T05:55:43.262619 | Meg | pythondev_help_Meg_2017-07-20T05:55:43.262619 | 1,500,530,143.262619 | 86,431 |
pythondev | help | furthermore, if you use vagrant or docker, you can make onboarding and environment issues between developers much smoother | 2017-07-20T05:56:17.275128 | Meg | pythondev_help_Meg_2017-07-20T05:56:17.275128 | 1,500,530,177.275128 | 86,432 |
pythondev | help | because _everyone_ is working from the same environment | 2017-07-20T05:56:26.278494 | Meg | pythondev_help_Meg_2017-07-20T05:56:26.278494 | 1,500,530,186.278494 | 86,433 |
pythondev | help | no worries about Windows issues when 3/4s of your team are on Linux | 2017-07-20T05:56:39.283190 | Meg | pythondev_help_Meg_2017-07-20T05:56:39.283190 | 1,500,530,199.28319 | 86,434 |
pythondev | help | <@Meg> Docker isn't for Windows? | 2017-07-20T06:08:03.531981 | Thomasina | pythondev_help_Thomasina_2017-07-20T06:08:03.531981 | 1,500,530,883.531981 | 86,435 |
pythondev | help | you misunderstand me | 2017-07-20T06:10:01.572126 | Meg | pythondev_help_Meg_2017-07-20T06:10:01.572126 | 1,500,531,001.572126 | 86,436 |
pythondev | help | docker is cross platform, and isolates your codebase from the environment | 2017-07-20T06:10:16.577534 | Meg | pythondev_help_Meg_2017-07-20T06:10:16.577534 | 1,500,531,016.577534 | 86,437 |
pythondev | help | <https://docs.docker.com/docker-for-windows/> | 2017-07-20T06:10:47.588094 | Meg | pythondev_help_Meg_2017-07-20T06:10:47.588094 | 1,500,531,047.588094 | 86,438 |
pythondev | help | <@Meg> Oh, yes, I just didn't understand your last sentence I guess. Thanks for telling me about Docker. For some reason I thought Docker was another PaaS or something. Is it actually meant to compliment other PaaS? | 2017-07-20T06:26:30.895568 | Thomasina | pythondev_help_Thomasina_2017-07-20T06:26:30.895568 | 1,500,531,990.895568 | 86,439 |
pythondev | help | Anyone here familiar with PostGIS? | 2017-07-20T07:31:30.078840 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:31:30.078840 | 1,500,535,890.07884 | 86,440 |
pythondev | help | I've done a bit with it. | 2017-07-20T07:33:57.122717 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:33:57.122717 | 1,500,536,037.122717 | 86,441 |
pythondev | help | <@Gabriele> Cool! Do you know where I might be able to get my hands on some sample geography data to experiment with? I have a database but it's full of shape and geometry data and I want to try querying for all objects within a certain radius of a point which I can't do without geography data. | 2017-07-20T07:37:00.177605 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:37:00.177605 | 1,500,536,220.177605 | 86,442 |
pythondev | help | you should be able to convert from geometry to geography | 2017-07-20T07:39:44.227012 | Meg | pythondev_help_Meg_2017-07-20T07:39:44.227012 | 1,500,536,384.227012 | 86,443 |
pythondev | help | <https://gis.stackexchange.com/questions/18011/is-it-possible-to-convert-from-geography-to-geometry-in-postgis> | 2017-07-20T07:39:57.231032 | Meg | pythondev_help_Meg_2017-07-20T07:39:57.231032 | 1,500,536,397.231032 | 86,444 |
pythondev | help | and furthermore, regarding docker
so, docker basically puts a bare bones VM in which you specify containers for parts of your application stack. | 2017-07-20T07:40:33.242022 | Meg | pythondev_help_Meg_2017-07-20T07:40:33.242022 | 1,500,536,433.242022 | 86,445 |
pythondev | help | so, if you're on windows and another dev is on mac | 2017-07-20T07:40:47.246412 | Meg | pythondev_help_Meg_2017-07-20T07:40:47.246412 | 1,500,536,447.246412 | 86,446 |
pythondev | help | <@Thomasina> I personally just used data from sources like Google Maps, querying for places and their long/lat, and adding to my DB | 2017-07-20T07:40:55.248623 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:40:55.248623 | 1,500,536,455.248623 | 86,447 |
pythondev | help | you can ensure that whatever issues you have are code related and not OS related | 2017-07-20T07:41:01.250420 | Meg | pythondev_help_Meg_2017-07-20T07:41:01.250420 | 1,500,536,461.25042 | 86,448 |
pythondev | help | furthermore, bringing in a new dev and getting them integrated with the projects is much simpler, because you already hhave the environment and code _right there_ | 2017-07-20T07:41:36.261085 | Meg | pythondev_help_Meg_2017-07-20T07:41:36.261085 | 1,500,536,496.261085 | 86,449 |
pythondev | help | <@Gabriele> Through the Google Maps API? Or do they have downloadable datasets?
<@Meg> Thanks! I'll look into that. Also, Docker does sound really useful. I'm going to check them out! Thanks! :smile: | 2017-07-20T07:42:12.271846 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:42:12.271846 | 1,500,536,532.271846 | 86,450 |
pythondev | help | through the API | 2017-07-20T07:42:24.275528 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:42:24.275528 | 1,500,536,544.275528 | 86,451 |
pythondev | help | <@Thomasina> <@Gabriele> <http://workshops.boundlessgeo.com/postgis-intro/geography.html#casting-to-geometry> | 2017-07-20T07:42:27.276409 | Meg | pythondev_help_Meg_2017-07-20T07:42:27.276409 | 1,500,536,547.276409 | 86,452 |
pythondev | help | if you already have the data, no need to use google maps | 2017-07-20T07:42:37.279183 | Meg | pythondev_help_Meg_2017-07-20T07:42:37.279183 | 1,500,536,557.279183 | 86,453 |
pythondev | help | and the US Census bureau already has plenty of data for download | 2017-07-20T07:43:15.290456 | Meg | pythondev_help_Meg_2017-07-20T07:43:15.290456 | 1,500,536,595.290456 | 86,454 |
pythondev | help | <https://www.census.gov/geo/maps-data/data/tiger.html> | 2017-07-20T07:43:16.290593 | Meg | pythondev_help_Meg_2017-07-20T07:43:16.290593 | 1,500,536,596.290593 | 86,455 |
pythondev | help | Yeah, I wasn't suggesting he use Google Maps, just saying that my application did | 2017-07-20T07:43:34.296243 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:43:34.296243 | 1,500,536,614.296243 | 86,456 |
pythondev | help | ah ok | 2017-07-20T07:43:42.298367 | Meg | pythondev_help_Meg_2017-07-20T07:43:42.298367 | 1,500,536,622.298367 | 86,457 |
pythondev | help | <@Gabriele> Oh, sorry I suppose I actually misunderstood my problem. I'm running this query:
`SELECT * FROM towns WHERE ST_Distance_Sphere(geom, ST_MakePoint(71.3033,44.2706)) <= 150 * 1609.34`
and the error is:
```ERROR: BOOM! Could not generate outside point!
CONTEXT: SQL function "st_distance_sphere" statement 2```
By `statement 2` I think it's actually referring to `ST_MakePoint(71.3033,44.2706)`. This is a valid lat/long though and looking at the documentation it appears `ST_MakePoint` does in fact accept a lat/long. :thinking_face: | 2017-07-20T07:49:19.400704 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:49:19.400704 | 1,500,536,959.400704 | 86,458 |
pythondev | help | <@Meg> Oh, not sure why census data never occurred to me. Nice! | 2017-07-20T07:49:52.410894 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:49:52.410894 | 1,500,536,992.410894 | 86,459 |
pythondev | help | you're in canada, right? there should be something similar for you there | 2017-07-20T07:50:09.416444 | Meg | pythondev_help_Meg_2017-07-20T07:50:09.416444 | 1,500,537,009.416444 | 86,460 |
pythondev | help | <http://www12.statcan.gc.ca/census-recensement/2011/geo/bound-limit/bound-limit-2016-eng.cfm> | 2017-07-20T07:51:02.433421 | Meg | pythondev_help_Meg_2017-07-20T07:51:02.433421 | 1,500,537,062.433421 | 86,461 |
pythondev | help | Yep. Although any dataset with a lot of records is good. I think I used the US census site awhile ago when doing machine learning so I'm a bit familiar with it, so I'd probably just use that again. | 2017-07-20T07:51:06.434827 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:51:06.434827 | 1,500,537,066.434827 | 86,462 |
pythondev | help | Oh, well since I don't have to look for it I might use the Canadian information then. :joy: Thanks | 2017-07-20T07:51:29.441934 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:51:29.441934 | 1,500,537,089.441934 | 86,463 |
pythondev | help | What I'm trying to do is get all recorded locations within a radius of a specific lat/long. I'm using PostGIS now which has been nice so far, but I'm wondering if there might be an even simpler solution... | 2017-07-20T07:53:16.475244 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:53:16.475244 | 1,500,537,196.475244 | 86,464 |
pythondev | help | and regarding your query | 2017-07-20T07:53:21.476776 | Meg | pythondev_help_Meg_2017-07-20T07:53:21.476776 | 1,500,537,201.476776 | 86,465 |
pythondev | help | what is `geom` | 2017-07-20T07:53:27.478768 | Meg | pythondev_help_Meg_2017-07-20T07:53:27.478768 | 1,500,537,207.478768 | 86,466 |
pythondev | help | To get all locations within a radius of somewhere is exactly the right sort of problem for PostGIS | 2017-07-20T07:53:56.487739 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:53:56.487739 | 1,500,537,236.487739 | 86,467 |
pythondev | help | yeah | 2017-07-20T07:54:02.489495 | Meg | pythondev_help_Meg_2017-07-20T07:54:02.489495 | 1,500,537,242.489495 | 86,468 |
pythondev | help | because you're working with spatial data | 2017-07-20T07:54:09.491617 | Meg | pythondev_help_Meg_2017-07-20T07:54:09.491617 | 1,500,537,249.491617 | 86,469 |
pythondev | help | The only simpler solution is to call someone else's API to perform the same operation on their database | 2017-07-20T07:54:13.492812 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:54:13.492812 | 1,500,537,253.492812 | 86,470 |
pythondev | help | and postgis is highly optimized | 2017-07-20T07:54:21.495510 | Meg | pythondev_help_Meg_2017-07-20T07:54:21.495510 | 1,500,537,261.49551 | 86,471 |
pythondev | help | you could do a naive solution | 2017-07-20T07:54:27.497078 | Meg | pythondev_help_Meg_2017-07-20T07:54:27.497078 | 1,500,537,267.497078 | 86,472 |
pythondev | help | but it would be significantly less performant | 2017-07-20T07:54:35.500256 | Meg | pythondev_help_Meg_2017-07-20T07:54:35.500256 | 1,500,537,275.500256 | 86,473 |
pythondev | help | `geom` is a column containing geometry data (in this case it's the boundaries of towns in a small section of the US). | 2017-07-20T07:54:40.501642 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:54:40.501642 | 1,500,537,280.501642 | 86,474 |
pythondev | help | <https://postgis.net/docs/ST_DistanceSphere.html> | 2017-07-20T07:54:55.506551 | Meg | pythondev_help_Meg_2017-07-20T07:54:55.506551 | 1,500,537,295.506551 | 86,475 |
pythondev | help | It needs points, not just any arbitrary geometry | 2017-07-20T07:55:08.510442 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:55:08.510442 | 1,500,537,308.510442 | 86,476 |
pythondev | help | Well I better stick with PostGIS then. While load will be very light at first it could become quite heavy later on. | 2017-07-20T07:55:17.513378 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:55:17.513378 | 1,500,537,317.513378 | 86,477 |
pythondev | help | for example, make a multilinestring | 2017-07-20T07:55:19.514163 | Meg | pythondev_help_Meg_2017-07-20T07:55:19.514163 | 1,500,537,319.514163 | 86,478 |
pythondev | help | <@Gabriele> I assumed the geometry was in projectable coordinates...? | 2017-07-20T07:56:04.528816 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:56:04.528816 | 1,500,537,364.528816 | 86,479 |
pythondev | help | <@Thomasina> if you want more help about gis stuff | 2017-07-20T07:56:13.531702 | Meg | pythondev_help_Meg_2017-07-20T07:56:13.531702 | 1,500,537,373.531702 | 86,480 |
pythondev | help | <http://thespatialcommunity.org/> | 2017-07-20T07:56:14.531843 | Meg | pythondev_help_Meg_2017-07-20T07:56:14.531843 | 1,500,537,374.531843 | 86,481 |
pythondev | help | It needs one single point | 2017-07-20T07:56:30.537323 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:56:30.537323 | 1,500,537,390.537323 | 86,482 |
pythondev | help | <@Meg> Oh nice! Didn't think there'd be a Slack for this. I requested an invite now. | 2017-07-20T07:57:12.550859 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:57:12.550859 | 1,500,537,432.550859 | 86,483 |
pythondev | help | are you using flask or django for this | 2017-07-20T07:57:56.565161 | Meg | pythondev_help_Meg_2017-07-20T07:57:56.565161 | 1,500,537,476.565161 | 86,484 |
pythondev | help | <@Gabriele> Oh yeah, my bad. I guess what I want to do then (to keep things simple) is to work out the center-point of each shape first. | 2017-07-20T07:58:01.566681 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:58:01.566681 | 1,500,537,481.566681 | 86,485 |
pythondev | help | I think there are other functions that are more lenient on which types they accept for distance calculations, but the docs for that one suggest that it doesn't | 2017-07-20T07:58:10.569479 | Gabriele | pythondev_help_Gabriele_2017-07-20T07:58:10.569479 | 1,500,537,490.569479 | 86,486 |
pythondev | help | <@Meg> Flask | 2017-07-20T07:58:10.569544 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:58:10.569544 | 1,500,537,490.569544 | 86,487 |
pythondev | help | <https://geoalchemy-2.readthedocs.io/en/latest/> | 2017-07-20T07:58:37.578114 | Meg | pythondev_help_Meg_2017-07-20T07:58:37.578114 | 1,500,537,517.578114 | 86,488 |
pythondev | help | might want to use that instead of raw queries, especially now | 2017-07-20T07:58:47.581818 | Meg | pythondev_help_Meg_2017-07-20T07:58:47.581818 | 1,500,537,527.581818 | 86,489 |
pythondev | help | <@Meg> Oh, yep, you or <@Gabriele> (or somebody at least) posted that here the other day so I have it. I'm running these queries raw in pgAdmin against the database though just to familiarize myself without having to run a .py script each time. :slightly_smiling_face: | 2017-07-20T07:59:58.604585 | Thomasina | pythondev_help_Thomasina_2017-07-20T07:59:58.604585 | 1,500,537,598.604585 | 86,490 |
pythondev | help | there isn't a shell for flask, like there is for django? | 2017-07-20T08:00:22.614295 | Meg | pythondev_help_Meg_2017-07-20T08:00:22.614295 | 1,500,537,622.614295 | 86,491 |
pythondev | help | There is | 2017-07-20T08:00:33.617840 | Thomasina | pythondev_help_Thomasina_2017-07-20T08:00:33.617840 | 1,500,537,633.61784 | 86,492 |
pythondev | help | so why not use that ORM instead? | 2017-07-20T08:00:54.625288 | Meg | pythondev_help_Meg_2017-07-20T08:00:54.625288 | 1,500,537,654.625288 | 86,493 |
pythondev | help | but my database in my Flask app isn't set up with the proper data | 2017-07-20T08:00:55.625819 | Thomasina | pythondev_help_Thomasina_2017-07-20T08:00:55.625819 | 1,500,537,655.625819 | 86,494 |
pythondev | help | gotcha | 2017-07-20T08:00:59.627362 | Meg | pythondev_help_Meg_2017-07-20T08:00:59.627362 | 1,500,537,659.627362 | 86,495 |
pythondev | help | I just loaded a sample dataset into a database in my local postgres install to play with it, for now | 2017-07-20T08:01:23.635559 | Thomasina | pythondev_help_Thomasina_2017-07-20T08:01:23.635559 | 1,500,537,683.635559 | 86,496 |
pythondev | help | Once I feel comfortable with how it works and the data I need and how to structure the tables for my data I'll do that part (hopefully soon!) | 2017-07-20T08:01:59.647804 | Thomasina | pythondev_help_Thomasina_2017-07-20T08:01:59.647804 | 1,500,537,719.647804 | 86,497 |
pythondev | help | alright | 2017-07-20T08:02:11.652181 | Meg | pythondev_help_Meg_2017-07-20T08:02:11.652181 | 1,500,537,731.652181 | 86,498 |
pythondev | help | So I've loaded a table with a list of world cities but I'm not getting the expected results. For example I can run `SELECT * FROM cities WHERE ST_DWithin(geog, ST_MakePoint(45.4215,75.6972)::geography, 470000)` which _should_ return Toronto and a number of other surrounding cities. I've confirmed that Toronto exists in the database with the expected lat/long. However 0 rows are returned. :disappointed: | 2017-07-20T09:56:37.852226 | Thomasina | pythondev_help_Thomasina_2017-07-20T09:56:37.852226 | 1,500,544,597.852226 | 86,499 |
pythondev | help | I'm not sure what I'm doing wrong. I've been following examples from articles and there are no errors returned...tried a few variations too. | 2017-07-20T09:58:59.941907 | Thomasina | pythondev_help_Thomasina_2017-07-20T09:58:59.941907 | 1,500,544,739.941907 | 86,500 |
pythondev | help | dear friendssss | 2017-07-20T09:59:38.967243 | Kandis | pythondev_help_Kandis_2017-07-20T09:59:38.967243 | 1,500,544,778.967243 | 86,501 |
pythondev | help | why does python tell me: | 2017-07-20T10:00:23.998438 | Kandis | pythondev_help_Kandis_2017-07-20T10:00:23.998438 | 1,500,544,823.998438 | 86,502 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.