error, I look forward to your GPL competitor.  Meanwhile, for anyone who wants to customise a public Abe instance without using a Github fork, I've added the following To-Do item:

 * Optionally add a link to each page allowing the user to download the
   source code and supporting files as a tarfile created on the fly to
   satisfy the AGPL.

I won't make this automatic due to security concerns, but I intend to make the new link simply tar up the directory containing abe.py, minus *.pyc and .git, and stream it to the browser as application/x-gtar-compressed or similar.  By turning this on and keeping your changes in the directory, you will satisfy the AGPL.  [Edit: or I will weaken Abe's license so this satisfies it.]  If you implement this before I get to it, feel free to submit a patch or pull request.

I'll also allow config file nesting so that secret configuration such as a database password can be left out.

n0m4d, thank you.

Confucius, you are welcome to use the URL in the OP, but do not expect as consistent service as you would get from BlockExplorer.com.  I run the service for demonstration purposes, the server can not handle BBE's load, the code is not yet as optimal as BBE's, and I will probably move the site to another host/port soon.

Cheers,
John

Your C++ looks all right to me.

I'm imagining a world where there are 10,000+ chains of which I'm interested in 100+.  Or maybe I'd like to run an exchange or a service similar to Block Explorer that by default supports any new chain on the net.  I imagine one p2p network, one port number shared by many or all chains.  Peers would exchange lists of chains that interest them so they can avoid useless relaying.  The protocol would support a "new chain" message type.  Of course, chains would need a unique identifier, perhaps a hash of some implementation of its block acceptance rules.

I'd also like a way to start hashing a chain that does not yet have a genesis block, only rules to validate the genesis block.  The rules might require, for example, the coinbase script to contain the hash of a particular (perhaps future) Bitcoin block as timestamp.  This would give people a fair chance at the genesis reward (if any) and permit a high difficulty for the block, especially in conjunction with shared proof of work.

Would you prefer to change things server-side (e.g., through templating) or would client-side scripting work?  If you can do it client-side, and you serve your scripts in unprocessed source form with an AGPL-compatible license notice, I won't object.  In fact, I'll adjust the output format to suit the task, e.g., XML for Ajax or HTML with data-centric class attributes.  And I'll add an optional <script> element to load your code.  I just want developers using your site to find it easy to grab your scripts, style sheets, etc., replicate their functionality by simply rehosting them, and extend them by editing them.

If you want to work server-side, I'd like to draw a line between functionality and presentation.  I'll be happy to include configurable static fragments at various points in the output, so long as these don't bring in proprietary applications that process chain data or scripts that fail to meet the above conditions.

Tell me your specific needs, and I will amend the license to meet them if compatible with my goals.  Want to change the CSS?  No problem.  Put in banner ads or change the donation address?  Sure.  Those things will be configurable when I get around to it.  But I'll have to be careful to exclude stuff that interacts with the data in ways that extend functionality, unless of course you are willing to fork the Github project or otherwise keep your changes public under the AGPL.

Or you could keep your site private and not worry.

Very interesting, thank you!

Do you plan to support, or would you consider patches to...

...share work with the BTC chain as described in Alternative Chains?

...one daemon multiple chains?

...chain discovery via IRC or other means?

...plug-in interface for block acceptance rule modules?

I would like Abe to support as many new chains as possible.  I'd prefer not to run a new daemon for each one, though different processes with one bitcoind executable is a step in the right direction.

2.7GB under /var/lib/postgresql/8.4/main

 pg_database_size
------------------
       2737561892

-rw-r----- 1 bitcoin  bitcoin  328428819 Jun 29 08:33 /home/bitcoin/.bitcoin/blk0001.dat
-rw-r--r-- 1 namecoin namecoin  11748764 Jun 29 08:11 /home/namecoin/.namecoin/blk0001.dat

This is with hashes and scripts stored as hex.  With a little work, I'm sure I could store them as binary.  And I hope to make the "coin-days destroyed" feature optional, which would save a bit of space for those who don't use it.  On the other hand, I plan to cache data in tables (optionally) for rendering web pages.  And I am considering storing addresses (rather than just public key hashes) for substring search.  So it could go up or down.

I'll do my best to match them.  /block, /tx, and /address are compatible, apart from some differences in output format and content.  /b doesn't work yet, because Abe currently lacks a notion of "default currency", and /b/NUMBER could be ambiguous.  (/chain/Bitcoin/b/NUMBER gives the equivalent.)  Other URLs could match BBE but aren't implemented yet.

I've just rebranded the project "Abe" (formerly "ABE") so as to reduce any chance of confusion with the original "BBE" (theymos' Bitcoin Block Explorer).

Correct.  The last thing I'd want is for Abe features to start showing up on a site like Block Explorer and growing closed-source enhancements.  But if someone will pay me for all the time I've put in, I will happily release it under a permissive license.  Or for a bit less, under an exclusive one.

For the sake of concreteness, as soon as 600BTC turn up at 1PWC7PNHL1SgvZaN7xEtygenKjWobWsCuf with 6 confirmations at difficulty over 1,000,000, you may have the current commit (8bd1c8609cab296174c68c7df06c60367c7b84c6) under the same license as bitcointools, bct-LICENSE.txt with my copyright next to Gavin's.  There are currently 0.05BTC, thanks to whoever.


Good point, and I should figure out which parts don't need AGPL protection, such as donation addresses and logos.


Added to the to-do list, thanks.  Please note that the database will also need optimising before it can support uses like SPV on any scale.  I have ideas for how to optimise it, but they are not at the top of my list yet.


My policy is that I sometimes look at my log files and get curious.    As for the software, it prints requested URLs to the standard output along with other information.  You may save or discard the output.

I've used Abe data to find some addresses that have had both BTC and NMC balances.  Example:

http://john-edwin-tobey.org:2750/address/128CYpCqrg9iWYLciBmV1FiQmS3bsPz3Jv

The Namecoin address with the same public key is MwhZkThpn4FH35b7z164DmsKVfSetrdwKe, but Abe shows the history of both at the above URL.  I consider this a "feature" not a "bug" though others may disagree  

Found in a database with BTC, NMC, and no other currencies using:

And used BlockExplorer's hashtoaddress function.  (Block Explorer is still good for some things... )

Announced under Project Development: http://forum.bitcoin.org/index.php?topic=22785.0

!!! Announcing Abe, the open-source block chain browser !!!

Newbies post here: https://bitcointalk.org/index.php?topic=51139.0

While theymos' Bitcoin Block Explorer has served us well, there are benefits that can only be had by an open-source version of this critical Bitcoin infrastructure.

• Privacy: you can view information without revealing your interest in it.
• Extensibility: you can add features that you want.
• Trust: you can audit the source code.
• Reliability: you can run it on your own server free of charge.

Abe provides much of Block Explorer's interface, including the list of recent blocks, a search tool, and detailed block, transaction, and address history pages.  Abe can also present multiple currencies such as Bitcoin, Namecoin, Tenebrix, and whatever FooCoin tomorrow may bring.  Abe lets you page back and forth in the block list and presents some new statistics like Average Coin Age.

Abe's interactive performance approaches Block Explorer's, though it is untested under BBE loads.  Abe supports a subset of Block Explorer's API.

Source code on Github: https://github.com/bitcoin-abe/bitcoin-abe
Demonstration site: http://abe.john-edwin-tobey.org/ (a 512MB VPS, can't handle heavy loads)
Requirements: Python and an RDBMS; tested with PostgreSQL, SQLite, MySQL, Oracle, ODBC, and DB2.
Donations: 1PWC7PNHL1SgvZaN7xEtygenKjWobWsCuf (BTC) and NJ3MSELK1cWnqUa6xhF2wUYAnz3RSrWXcK (NMC).

Enjoy!

December 6, 2012: Version 0.7.2 released.
August 31, 2011: Version 0.6 released.
August 16, 2011: Version 0.5 released.
July 15, 2011: See changes since July 4.

I guess.. 07: length of string that follows; 76657273696f6e: ASCII for "version"; 027d0000: little-endian version number 32002 (0.3.20.2); 0a: length of string that follows; 64656661756c746b6579: ASCII for "defaultkey"; 41: length of public key that follows.

Thanks!

Do you mean the "raw" block and transaction pages?  It's on the to-do, but not near the top.  Patches are welcome. 

Does any string of 65 bytes starting with 04 occur more than once?  Can you convert it to an address (e.g., with Block Explorer) and see a balance belonging to that address?  If you can get past these steps, you may have a chance.

Yes.  The block chain concept is brilliant, the code is very good, but the BTC "monetary policy" leaves much to be desired.  This is not a criticism of Satoshi, whose brilliant work leaves us a foundation to build upon.

My hope is to make it easy for anyone with an economic theory (not just C++ programmers) to start a block chain with a new policy, get it mined at high difficulty (leveraging the Bitcoin network), and promote the new currency.  To this end, I am developing Another Block Explorer with a focus on multiple currencies.  I would hope someone (perhaps I) creates an open-source exchange for Bitcoin-like currencies, a client with multi-currency support (including genesis block creation and pluggable block acceptance rules), a set of configurable policy modules, and a "currency designer" application.

Over the ages, kings and banks have tried many monetary policies with varying success.  I believe the path to a better one lies through experiment and a wealth of options.

Not sure if I made 1 month or you paid out already, but ABE now does this.  https://github.com/jtobey/bitcoin-abe  Here is your query:

select cc.block_height, b.block_nTime, 100 * (1 - b.block_satoshi_seconds / b.block_total_ss) percent_btcdd from chain_candidate cc join block b using (block_id) where cc.chain_id = 1 and cc.in_longest = 1 and b.block_total_ss > 0 order by cc.block_height;

About to be on http://john-edwin-tobey.org:2750/chain/Bitcoin/ as soon as the schema upgrade completes.  Though that site is "development" and I know of no "production" ABE site.

Well, I don't think he understands the beauty of the block chain.  The Bitcoin "bag holder" still has private keys and proof of a lot of work attesting to their value.

But the early Bitcoin adopters got too much.  Not "too much" in the sense of "unfair" nor in the sense of "I want them" or "it can't work", but enough to plague the currency with exchange rate volatility for the foreseeable future.  Perhaps even enough to limit the exchange value's upside.

Edit: latest features & discussion: https://bitcointalk.org/index.php?topic=22785

Tired of waiting for theymos to release the original Block Explorer code, just before khal's Namecoin explorer announcement, I went and started my own based on Gavin's bitcointools.

Code: https://github.com/jtobey/bitcoin-abe
Service: http://abe.john-edwin-tobey.org/ and anywhere you host it

June 26 updates in orange
Features:

• free and open source (GNU Affero General Public License)
• block, transaction, and address history pages similar to http://blockexplorer.com/
• block/tx/address search box
• does not require patching bitcoind (uses slightly modified bitcointools, not getblock)
• low block delay: often shows blocks not yet on http://blockexplorer.com/
• chain page can list more than the 20 newest blocks
• can handle several chains in the same database (Bitcoin, Testnet, Namecoin, ...)
• real-time Bitcoin Days Destroyed and Average Coin Age data
• written in Python and portable SQL, tested with PostgreSQL and Sqlite

Coming later, maybe:

• speedups to match http://blockexplorer.com/, I hope
• stats pages/API like http://blockexplorer.com/q
• see the TODO.txt file
• support for indirect block hashes along these lines
• Whatever you implement

Enjoy!