sv-dlp is a powerful API Wrapper for Street View platforms, as it simplifies downloading panoramic images, fetching metadata, and creating short URLs. sv-dlp offers extensive internal functionality, including metadata extraction and panorama stitching. It's highly adaptable, allowing developers to integrate it into their projects.
This project has been made for educational purposes and all content scraped from is owned by their respective authors.
For detailed documentation and usage examples, please refer to the official documentation.
Please note that sv-dlp relies on reverse-engineered APIs, so its functionality may depend on API stability. If you encounter issues, report them with the "Dead API" label for a prompt resolution. Additionally, respect service limitations, and rest assured that we're committed to keeping sv-dlp up to date and running smoothly for all supported platforms.
- Download Panoramas
- A panorama ID, a coordinate or a short URL can be parsed
- [Google] Radius can be parsed if input is coordinates
- Is multithreaded; each panorama tile row is in different threads for faster downloads
- Saves individual tiles if
--save-tilesis parsed - Download from .csv/.json files
- A panorama ID, a coordinate or a short URL can be parsed
- Fully print metadata
- Coords, Date, Panorama ID and Gen can be printed separately
- Coordinates and short URLs are automatically turned to Panorama IDs
- Short links with a Panorama ID or coordinates
| Service | Progress | Notes |
|---|---|---|
| 99% | Zoom 5 only obtains a portion of the face. | |
| Baidu | 99% | Translation between BD09MC and WGS84 is innacurate. |
| Naver | 99% | Short Link function does not use its appropiate API call. |
| Yandex | 99% | Short Link function does not use the shortenPath API call. |
| Apple | 95% | Misc features not supported; can only obtain metadata by coords - credit to sk-zk & retroplasma |
| Bing | 95% | Misc features not implemented; can only obtain metadata by coords - credit to sk-zk |
To install sv-dlp, download it via PIP, as long as Python >3.10 is installed
$ pip install sv_dlp
$ sv-dlpIf you prefer using a binary instead, check out the binaries here
and add them to your PATH folder. Check out this Windows Guide
for more information if you're not familiar with it.
sv_dlp's features are contained within a class for syntax convenience, so to begin import the module and declare a variable of the class
import sv_dlp
sv_dlp = sv_dlp.sv_dlp()- Whenever you call any function that is relating to downloading, or translating coordinates
to Panorama ID,
self.get_metadata()is called within that function, therefore it is allocated inself.metadata. If you plan on tinkering with the metadata, it is is recommended to callself_get_metadata()first before calling any other function for syntax readibility. - Metadata is returned as a
MetadataStructureobject. If you wish to use a dictionary instead, consider using the.dict()method - If you wish to call a function with a different service than google, please use
self.set_service()first before doing anything - input must be lowercase
To download panorama directly from coordinates:
pano_img = sv_dlp.download_panorama(lat=6.603079535799973, lng=-73.99819681137278)
sv_dlp.postdownload.save_panorama(pano_img, sv_dlp.metadata)To download panorama directly from Panorama ID:
pano_id = sv_dlp.get_pano_id(lat=6.603079535799973, lng=-73.99819681137278)
pano_img = sv_dlp.download_panorama(pano_id)
sv_dlp.postdownload.save_panorama(pano_img, sv_dlp.metadata)To download panorama directly from metadata:
metadata = sv_dlp.get_metadata(lat=6.603079535799973, lng=-73.99819681137278)
pano_img = sv_dlp.download_panorama(metadata.pano_id, zoom=max)
sv_dlp.postdownload.save_panorama(pano_img, metadata)To download tiles individually:
pano, tiles = sv_dlp.download_panorama("YV7i9WYmvPqT5nEtFLq3SA")
sv_dlp.postdownload.save_tiles(tiles, sv_dlp.metadata)You can also pass a single variable to sv_dlp.download_panorama,
as postdownload will automatically determine
if it holds the panorama and tiles.
To obtain older and linked panoramas from given location:
metadata = sv_dlp.get_metadata(lat=6.603079535799973, lng=-73.99819681137278, get_linked_panos=True)
for pano in metadata.timeline:
print(pano)
for pano in metadata.linked_panos:
print(pano)To obtain a panorama's date:
metadata = sv_dlp.get_metadata(lat=6.603079535799973, lng=-73.99819681137278)
date = metadata.date
print(date)Yes! You can use the .dict() method to retrieve the metadata as a dictionary
instead.
Depending on how close the given coordinates are, it'll download the same panorama ID no matter the difference. It's recommended to use .json or .csv with panorama IDs instead.
Depending on how accesible it is scraping-wise, yes. Feel free to submit in an issue about the specific service and I'll gladly see what I can do about it. If you're a developer though, refer to the next question.
Sure! You're welcome to submit in a pull request as long as its to improve this program, such as improving performance, fixing a bug, adding a service, etc. For those who want to add a service, check out the documentation for it.