Draft: Feat/glom like key searching #44
Conversation
| res = f"✨ Scenario: {'🔹' * self.readers}{'🔻' * self.writers} ({self.readers}r{self.writers}w)" | ||
| res += ", 🔸 compression" if self.use_compression else "" | ||
| res += ", 💎 big file" if self.big_file else "" |
There was a problem hiding this comment.
I will put this in a different PR to keep this one clean
| """ | ||
| The Indexer takes the name of a database file, and tries to load the .index file | ||
| of the corresponding database file. | ||
|
|
||
| The name of the index file is the name of the database file, with the extension | ||
| .index and all "/" replaced with "___" | ||
|
|
||
| The content of the index file is a json object, where the keys are keys inside | ||
| the database json file, and the values are lists of 5 elements: | ||
| - start_index: The index of the first byte of the value of the key in the database file | ||
| - end_index: The index of the last byte of the value of the key in the database file | ||
| - indent_level: The indent level of the key in the database file | ||
| - indent_with: The indent string used. | ||
| - value_hash: The hash of the value bytes | ||
| """ | ||
|
|
||
| __slots__ = ("data", "path") | ||
|
|
||
| def __init__(self, db_name: str): | ||
| # Make path of index file | ||
| db_name = db_name.replace("/", "___") | ||
| self.path = os.path.join(config.storage_directory, ".ddb", f"{db_name}.index") | ||
|
|
||
| os.makedirs(os.path.dirname(self.path), exist_ok=True) | ||
| if not os.path.exists(self.path): | ||
| self.data = {} | ||
| return | ||
|
|
||
| try: | ||
| with open(self.path, "rb") as f: | ||
| self.data = orjson.loads(f.read()) | ||
| except orjson.JSONDecodeError: | ||
| self.data = {} | ||
|
|
||
|
|
||
| def get(self, key): | ||
| """ | ||
| Returns a list of 5 elements for a key if it exists, otherwise None | ||
| Elements:[start_index, end_index, indent_level, indent_with, value_hash] | ||
| """ | ||
| return self.data.get(key, None) | ||
|
|
||
|
|
||
| def write(self, key, start_index, end_index, indent_level, indent_with, value_hash, old_value_end): | ||
| """ | ||
| Write index information for a key to the index file | ||
| """ | ||
|
|
||
| if self.data.get(key, None) is not None: | ||
| delta = end_index - old_value_end | ||
| for entry in self.data.values(): | ||
| if entry[0] > old_value_end: | ||
| entry[0] += delta | ||
| entry[1] += delta | ||
|
|
||
| self.data[key] = [start_index, end_index, indent_level, indent_with, value_hash] | ||
| with open(self.path, "wb") as f: | ||
| f.write(orjson.dumps(self.data)) |
There was a problem hiding this comment.
@UmbrellaMalware let's stick with tabs so the diff looks clean
There was a problem hiding this comment.
pep8 says spaces are preferred
This can become a problem if someone else wants to propose changes to the project
There was a problem hiding this comment.
Yes I know. I personally prefer tabs, and think they are faster to work with. But if this project becomes bigger, I agree that it should adhere to most python community standards. But I'd like to keep the merge clean, the indentation and formatting stuff should be part of a separate issue/MR
mkrd
force-pushed
the
feat/glom-like-key-searching
branch
from
25989d2
to
4d0c4b8
Compare
|
@UmbrellaMalware I did a quick rebase to incorporate die latest changes in main |
I need to test new functionality, I just found a bug in the search algorithm)) I'm also doing some documentation updates. |
Do you mean the issue with reading a key value that is an integer or float? |
No, if u try to read "Key.Subkey1.Subkey2" returns not found, but subkey2 exists |
mkrd
changed the title
Things left to do: