TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Show HN: Excel to Python Compiler

79 点作者 narush12 个月前
We (me and @aarondia) built a tool to help you turn psuedo-software Excel files into real-software Python. Ideally, Pyoneer helps you automate your manual Excel processes. You can try it today here: <a href="https:&#x2F;&#x2F;pyoneer.ai" rel="nofollow">https:&#x2F;&#x2F;pyoneer.ai</a>.<p><i>How it works:</i><p>1. You upload an Excel file<p>2. We statically parse the Excel file and build a dependency graph of all the cells, tables, formulas, and pivots.<p>3. We do a graph traversal, and translate nodes as we hit them. We use OpenAI APIs to translate formulas. There’s a bunch of extra work here — because even with the best prompt engineering a fella like me can do, OpenAI sucks at translating formulas (primarily because it doesn’t know what datatypes its dealing with). We augment this translation with a mapping from ranges to variable names and types, which in our experience can improve the percentage of correctly translatable formulas by about 5x.<p>4. We generate test cases for our translations as well, to make sure the Python process matches your Excel process.<p>5. We give you back a Jupyter notebook that contains the code we generated.<p>If there are pieces of the Excel we can’t translate successfully (complex formulas, or pivot tables currently), then we leave them as a TODO in the code. This makes it easy for you to hop in and continue finishing the script.<p><i>Who is this for:</i><p>Developers who know Python, primarily! Pyoneer might be useful if:<p>1. You’ve got an Excel file you’re looking to move to Python (usually for speed, size, or maintenance reasons).<p>2. There’s enough logic contained in the notebook that it’s going to be a hassle for you to just rewrite it from scratch.<p>3. Or you don’t know the logic that is in the Excel workbook well since you didn’t write it in the first place :)<p>Post translation, even if Pyoneer doesn&#x27;t nail it perfectly or translate all the formulas, you&#x27;ll be able to pop into the notebook and continue cleaning up the TODOs &#x2F; finish writing the formulas.<p><i>What the Alpha launch supports:</i><p>Launched early! Currently we’re focused on supporting:<p>1. Any number of sheets, with any reference structure between them.<p>2. Cells that translate as variables directly. We’ll translate the formulas to Python code that has the same result, or else we’ll generate a TODO letting you know we failed translating this cell.<p>3. Tables that translate as Pandas dataframes. We support at most one table per sheet, at the tables must be contigious. If the formulas in a column are consistent, then we will try and translate this as a single pandas statement.<p>We do not support: pivot tables or complex formulas. When we fail to translate these, we generate TODO statements. We also don’t support graphs or macros - and you won’t see these reflected in the output at all currently.<p><i>Why we built this:</i><p>We did YCS20 and built an open source tool called Mito(<a href="https:&#x2F;&#x2F;trymito.io">https:&#x2F;&#x2F;trymito.io</a>). It’s been a good journey since then - we’ve scaled revenue and to over 2k Github stars (<a href="https:&#x2F;&#x2F;github.com&#x2F;mito-ds&#x2F;mito">https:&#x2F;&#x2F;github.com&#x2F;mito-ds&#x2F;mito</a>). But fundamentally, Mito is a tool that’s useful for Excel users who wanted to start writing Python code more effectively.<p>We wanted to take another stab at the Excel -&gt; Python pain point that was more developer focused - that helped developers that have to translate Excel files into Python do this much more quickly. Hence, Pyoneer!<p>I’ll be in the comments today if you’ve got feedback, criticism, questions, or comments.

11 条评论

bko12 个月前
Why would you get an LLM to transpile the formula for you? There are already libraries that attempt to do this. Wouldn&#x27;t your time be better served in expanding these libraries to overcome deficiencies? I don&#x27;t get why you would throw away all the work done in this space and just offload it to an LLM. It&#x27;s a technical problem. Imagine if ditched all transpilers and gave it to an LLM and said &quot;good luck!&quot;<p><a href="https:&#x2F;&#x2F;formulas.readthedocs.io&#x2F;en&#x2F;stable&#x2F;" rel="nofollow">https:&#x2F;&#x2F;formulas.readthedocs.io&#x2F;en&#x2F;stable&#x2F;</a>
评论 #40459520 未加载
评论 #40489749 未加载
评论 #40459351 未加载
LeoPanthera12 个月前
&gt; We use OpenAI APIs to translate formulas.<p>It&#x27;s very uncool that you don&#x27;t make this <i>very</i> clear on the landing page.
评论 #40460150 未加载
评论 #40461861 未加载
评论 #40467910 未加载
rfergie12 个月前
<a href="https:&#x2F;&#x2F;pypi.org&#x2F;project&#x2F;formulas&#x2F;" rel="nofollow">https:&#x2F;&#x2F;pypi.org&#x2F;project&#x2F;formulas&#x2F;</a> might help with this without needing the OpenAI part
评论 #40459355 未加载
hk133712 个月前
This is really awesome.<p>&gt; In addition, Content found on or through this Service are the property of Saga Inc. or used with permission. You may not distribute, modify, transmit, reuse, download, repost, copy, or use said Content, whether in whole or in part, for commercial purposes or for personal gain, without express advance written permission from us.<p>Is the &quot;Content&quot; the results of what is uploaded?
评论 #40459296 未加载
flashgordon12 个月前
So is this a one-way only process (I&#x27;d assume it is)? This assumes that the once the python engineers who are &quot;tasked&quot; to productionize it, no more changes are allowed to the sheets or do you allow incremental updates?
评论 #40459669 未加载
jeffwiederkehr12 个月前
I stumbled into a career in programming due to originally learning excel in an unrelated field. This project is such a great idea imo and seems like it would be a game changer for anyone that is regularly working between excel and python files for data.<p>Also my take on the transpilation vs LLM is that you chose the right path. The point of transforming an excel spreadsheet to python is to have better organization and ability to reason about the operations being performed.
评论 #40460937 未加载
alok-g12 个月前
Suggesting an extreme test case for you:<p><a href="https:&#x2F;&#x2F;github.com&#x2F;amzn&#x2F;computer-vision-basics-in-microsoft-excel">https:&#x2F;&#x2F;github.com&#x2F;amzn&#x2F;computer-vision-basics-in-microsoft-...</a><p>(I&#x27;m the primary author of the above.)
gaze12 个月前
Sorry &quot;morgtage.xls&quot;? This makes me feel a bit pessimistic about the accuracy of the formulas.
评论 #40460171 未加载
trevzercap12 个月前
Very cool, what are your plans for working with external data sources via power query?
评论 #40459226 未加载
kingkongjaffa12 个月前
What customer discovery have you done so far?<p>&gt; turn psuedo-software Excel files into real-software Python<p>I&#x27;m curious how many people actually have this problem. It strikes me that the corporate environments that are building big hairy excel files probably have locked down IT where running python is a difficult thing to do.<p>In the environments where you can get some kind of python running, it&#x27;s probably likely that developers are looped in as part of an efficiency project, and doing some kind of voice of the customer &#x2F; requirements capture and starting to build from scratch more or less.<p>There&#x27;s a vanishing window for stuff like this, if you&#x27;re a Microsoft shop like 99% of the corporate world I think you are turning those excel files into power apps and powerBI dashboards, before you are hiring python devs.
评论 #40459050 未加载
评论 #40458792 未加载
评论 #40458807 未加载
评论 #40460252 未加载
评论 #40458756 未加载
imaddjeddai12 个月前
I&#x27;ll give it a try! Thanks!