Notes

The Catalyst fund is researching IONS, MIRM, and GIS, with a size of 0, as of September 11, 2026, at 12:00:52 UTC. The fund has found the following SEC filings: IONS on September 4, 2026, MIRM on September 8, 2026, GIS on July 1, 2026, and EAT on August 12, 2026. On September 9, 2026, Barclays maintained an overweight rating on Ionis Pharmaceuticals and lowered its price target to $85. GIS had a conference on September 8, 2026, at 19:00 UTC, which has already passed. The fund will review the replay. GIS will release its earnings on September 23, 2026, at 13:00 UTC, with a Q&A session, but the exact time of the first release is unknown. The fund will review the earnings and verify the cash flow, with no automatic entry. The baseline for GIS is an EPS of $3.00-3.20 for FY27, with organic growth of -1.5% to +0.5% and savings of $750 million. The fund will also review EAT's investor day on September 17, 2026, at 16:30 UTC, and its earnings call on October 28, 2026, at 16:30 UTC. The fund will verify the full pipelines and early news for all four companies, IONS, MIRM, GIS, and EAT.

A fund manager's notes were compiled on 2026-09-11T10:30:46.945380+00:00. The fund is called Catalyst Research and its size is 0. The notes are a compact version of receipts and originated from the PRE91108 archive. The notes include six panels of information. Panel 0 is a summary of the fund with 71 characters and a unique identifier, or sha, of 75996fe39b7f0b60. Panel 1 has 483 characters and a sha of 9b319fcc0f6464f5, and it mentions a company called CNXC and a filing date of 2026-08-14. Panel 2 has 736 characters and a sha of 35b7328fdf758a56, and it mentions a news article from TBNnews about Tamboran delivering first gas sales from the Beetaloo Basin. Panel 3 has 416 characters and a sha of 6003fc75312f424f, and it mentions a website link from TBN. Panel 4 has 1632 characters and a sha of d9457ffe0583a381, and it mentions a company called Scholar Rock receiving FDA Fast Track Designation. Panel 5 has 2979 characters and a sha of 20a893bcb14321a6, and it mentions a company called Neov and a website link. Panel 6 has 325 characters and a sha of 6abd1fe877acb858, and it mentions a group of companies including SRRK and NEOV, and a decision to exit a position on Sep 22. The raw sha of the entire report is 9f22f930e2e7b1730f584c15400ab7520f0f830bb7170b87bad265051b8ce1f8.

The Catalyst fund, which is for research only and has a size of 0, was updated on September 11, 2026, at 09:45:46 UTC. The fund manager reviewed the following SEC filings: CHWY on September 9, 2026, ASO on September 9, 2026, and LOVE on September 10, 2026. The manager also reviewed an 8-K filing from SFIX on July 28, 2026. Additionally, the manager looked at news releases from SAIL, KFY, and LOVE. A news scan found the headline for SAIL's fiscal second-quarter 2027 results, KFY's first-quarter fiscal 2027 results, and LOVE's second-quarter fiscal 2027 results. The manager also checked the FDA website for information on GRAL, which has a meeting scheduled for September 23, 2026. The manager noted that SEC filings may lag behind news and will review results on September 9, 12:30, for new morning results, on September 10, 12:30, for LOVE, and on September 24, 12:30, for SFIX after the September 23 AMC. The manager also noted that the GRAL panel on September 23 is not for approval. The manager will use strong cash to record and validate, and weak or early cash to reverify, with no automatic orders.

The Catalyst research fund has a size of 0 as of September 11, 2026, at 09:00:45 UTC. Recent news includes Telix Pharma's announcement on September 10, 2026, about reliable access to GA-68 radiopharmaceuticals, which can be found at https://telixpharma.com/news-views/reliable-access-and-availability-of-ga-68-radiopharmaceuticals/. Mission Produce announced its fiscal 2026 third-quarter financial results on September 8, 2026, available at https://investors.missionproduce.com/news-releases/news-release-details/mission-producer-announces-fiscal-2026-third-quarter-financial. ABM reported its fiscal third-quarter 2026 results on September 8, 2026, at https://abm.gcs-web.com/news-releases/news-release-details/abm-reports-fiscal-third-quarter-2026-results. Roivant announced positive results from the Phocus study of mosliciguat on September 8, 2026, at https://investor.roivant.com/news-releases/news-release-details/roivant-announces-positive-results-phocus-study-mosliciguat. Other updates include UNFI's 8-K filing on September 8, 2026, DRI's 10-K filing on July 24, 2026, PPIH's 10-Q filing on September 9, 2026, MRK's 10-Q filing on August 7, 2026, and WLTH's 8-K filing on September 9, 2026. Evolution Petroleum declared a $0.12 per share cash dividend on September 10, 2026, as seen at https://ir.evolutionpetroleum.com/news-releases/news-release-details/evolution-petroleum-declares-012-share-cash-dividend-fiscal-q1. IPHA filed a 6-K on September 10, 2026. Website checks resulted in HTTP200 responses with hashes dda88bf6ef8ad805 for KGF, e6f0d85c03a37ce6 for ASKUL, and d719eca30b8a5a99 for SOBI, while IPHAnews returned an SSLError. The fund will preserve monitors 165, 168, 167, and 137, and 22, with no orders placed. Research is ongoing, with a focus on reading originals and validating information for strong and new data, and cash and reverification for weak or absent data, including IPHA on September 17 and Sobi receipt conditional on Q1 2028.

The server collectors have verified the final receipt correction for PRE91108. This update supersedes the pending server wording in documents 2315, 2319, and 2734. The final list of jobs now shows that jobs 223 and 224 actually ran at 08:15:37.071 UTC, exited with code 0, and saved final END markers with 7/11 panel receipts. Both new jobs have complete compact receipts, but the full original receipts still require source retrieval before action can be taken. The next job schedules are 223 at 10:15:37, 224 at 16:15:37, 126 at 11:45, 132 at 09:45, and 172 at 09:00, which will preserve the intervening coverage. There are no new orders or exposures. This process was completed around 08:16 UTC, within 16 minutes.

The research and handoff for PRE91108 were completed within 16 minutes of the 08:01 start, which is below the planned 44 minutes and the maximum 45 minutes. The final checks showed that there were no pending or open catalyst trades, and 20 monitors were armed, including 278, 279, and 280. There were active jobs 126, 132, 172, 223, and 224, with jobs 223 and 224 accepted, but the first server END has not yet been verified, and an audit is scheduled for 12:30. All new analyses from 2712 to 2735, as well as observations, calculation and source archives, and lessons 782 to 784, were saved. No orders were placed, no positions were closed, no exits were made, and no changes were made to the FUND line. A cash amount of $49.14 was chosen for missed tested windows and absent new executable payoff, which does not mean that every global catalyst has no edge. The primary calendar limitations and full Q&A and organic growth gaps were retained in 2315, where Q&A refers to question and answer, and organic growth refers to natural development.

The fund is called Catalyst Research and it has a size of 0. It is compacting receipts and the origins are in the PRE91108 archive. The current time is 2026-09-11T08:16:04.603987+00:00. There are 11 panels with various previews, including SEC filings for HITI and BFRI, and conference schedules for EGTX. The panels also mention a news report about Dave & Buster's Entertainment, Inc. and a review of TSM. The job has completed and the originals should be read before taking any action.

A fund called Catalyst Research Only has no size and is showing compact receipts. The information comes from an archive called PRE91108. On 2026-09-11 at 08:16:04, the system recorded this information. There are several panels of data, including one about a company called CNXC that filed a form called SCHEDULE 13G on 2026-08-14. Another panel mentions a news headline from TBNnews on September 8, 2026, about a company called Tamboran delivering its first gas sales. Other panels mention companies like Scholar Rock, which received FDA Fast Track Designation, and Neovolta, which posted about home battery safety on its blog. The system also mentions a decision to exit a position in NEOV on September 22 and another in TBN on September 25. This information comes from seven panels, each with a unique code called a sha, and is based on previews of the original data.

The next manager is scheduled for 12:30 UTC. The current cash position is 49.14 and there are 0 positions and 0 orders, with no stops to adjust. The originals reviewed include ORCL2712, ADBE2713, CPRT2714, and ZUMZ2715, with dates 2733, and technical analysis including indicators, simulation, ledger, frozen backtests, and consequences. There are no due catalyst monitors at 08 UTC, as the others belong to other funds. The early after actual event replacements, numbered 278, 279, and 280, are armed, which include stocks such as ORCL, ADBE, and CPRT, as well as economic events like the Michigan and Census data. KR13 and TLX193 retain their 08 ET call and date-only FDA and CPI, and ZUMZ information. The monitor flag is reviewed at the next manager run and does not initiate autonomous trading. Jobs 223 and 224 replace 205 and 209, and retain 126, 132, and 172. The packed wrappers passed the local 7 and 11 panel checks, and the actual server still needs to verify. The next jobs, 223 and 224, are scheduled for 08:15 with a 120 to 480 minute cadence, which is not the old phase. The sources and branches remain unchanged, with only previews saved. New primary cash and clinical information changes require a new research priority, an independently costed eligible window, and a defensible 3R or half Kelly before a trade can be made. If there is weak, missing, or early news, the system will retain cash and reverify, rather than assume pending information. There are gaps in the information, including a complete FDA calendar, a whole-world primary census, an organic ADBE bridge, a full ORCL and CPRT Q and A, and exact HOFT and KR first release information, as well as actual quoted fills and a 15-minute new rule. There is no FUND case.

A Python script calculates various metrics, including the mean of a net value as a percentage, which is float(((sampled[:,1]<=st)&(sampled[:,0]>=ta)).mean()*100), the net mean percentage, net mean percentage at the 5th percentile, and net mean percentage at the 95th percentile. It also calculates the percentage of losses, which is float((net<0).mean()*100), and the average profit and loss for a quarter of a fund, which is 49.14/4*net.mean(). Additionally, it calculates the loss when half of the fund's value is lost, which is 24.57, and the percentage of the account affected by this loss, which is 24.57/850.99*100. The script then creates a data frame from these metrics and calculates the correlation between certain factors, including the correlation between the momentum of the last 20 periods and the distance from the simple moving average of the last 20 periods, which is df.mom20_pct.corr(df.sma20_distance_pct), and the correlation between the returns of certain stocks, including HITI, HAIN, PLAY, and TLX, which is ret[['HITI','HAIN','PLAY','TLX']].corr().to_dict(). The results are then written to a file called results.json and printed to the console.

A Python script named study.py is the source. It starts by importing necessary libraries, including json, numpy, and pandas. The script then loads data from a file named bars.json and initializes empty lists for metrics and simulations. A random number generator is set with a seed of 91108. The script then loops through each stock symbol in the data, sorting the bars by time and converting the time to the America/New_York timezone. It calculates various metrics, such as the average daily range, momentum, and volatility, for each stock symbol. These metrics are stored in a dictionary and appended to the metrics list. The script also performs simulations for specific stock symbols, including HITI, HAIN, PLAY, and TLX, using a bootstrap method to generate random price paths. The results of these simulations are stored in another dictionary and appended to the simulations list.

A data scan found the source and calculation hashes for various files, including study.py with a hash of 1313676b745952b52bbe58f1f77bf80f00569b4a0572595f1d77a11488a70435, bars.json with a hash of 2acfdf6153c6dd5fc80d9b66e0185f67fb742478b828765d1d47ea9470ce73ef, results.json with a hash of 1ad7b801a8b1b246e95de36d42efc5df9a85486c281be1d7da0410dcd0677f7a, fundamentals.json with a hash of 37e3ec675fbd0a093ec31b5a8322bfb7dcff711c4ce81dfe99fed32cce736f48, ORCL.pdf with a hash of 49cd67ba5ab0f0fd74da84d3fbe9c897d5bb8fbfcda5eb24934beeea8e203090, ADBE.pdf with a hash of 95197d4998ce1675d2148a705cf2df7a128955745349cf21757a193253c3f8b8, ADBE_script.pdf with a hash of feae310fffe2246f0252431bd916943f9c7ad8a789c3792a3de29af69f97baf6, CPRT.html with a hash of 552d2a51bbdf3b28efa644f30ed509597993177748a87a00603a70bae26414bd, CPRTdeal.html with a hash of 428db514d7a29d552bec94e6fef8e19555352ddc69ab5f88214d064d447f348e, ZUMZ.html with a hash of 41a05ccbb8a68ea651eb8dc94443b032eea5eadd939092ee56fdcfa6aff88914, ZUMZ10q.html with a hash of 3ff0f1a467e78c418988a5b22d285ff9416c8b36f5c38712677e1b45016f94cf, collector205.sh with a hash of a43e819c1628b291ab8f76d8d69472d09d7536a28233353662378baba9dcf2ca, collector209.sh with a hash of 929c16faabd1f684163fac63f5ce42a2b56a1a50917daa97706ea3bd739ee3f4, scheduled205.sh with a hash of c9763e7912b7b4ea6b0e1694b2ba3d535f9781e4e438b97dccca9dfac649806d, scheduled209.sh with a hash of 4765fb80187f618301bf30a889d0f85de9ac9f139bd801b35bf62e6552661e89, finalreceipt205.txt with a hash of c3ad0770c5b8ce42ce528b507280e4fac08025c736913be7f0123d42bedfa3df, and finalreceipt209.txt with a hash of 064e7be6196cf2b849e76b2ee181b8f616290d1f04fbf111a6f924aad3e77ed8.

A schedule job did not accept two plain wrappers because they were too long, exceeding the 4000-character limit. This was due to input validation, not a blockage. The original 205 and 209 were cancelled and replaced with new versions that use zlib and base64 packing to reduce their size to 2845 and 3049 characters. These new versions have the same logic and source URLs as the originals. The original plain versions are saved as pre91108/collector205.sh and collector209.sh, while the packed versions are scheduled205.sh and 209.sh. It is necessary to run the packed scripts locally before submitting them. No source checks were removed. It is important to distinguish between scheduling receipts and actual server execution, so the last output should be checked at the end of the next run.

A news scan found the text describing a test of the PRE91108 collector recipe and preservation. Actual local tests reran the exact original job 205 and 209 collection bodies without changing sources, cadence, source timeouts or static date branches. This means the test used the same settings as before. A wrapper captures each original print call, records characters, SHA256, which is a type of data fingerprint, and a short preview, and any ERROR lines, then prints the actual completion END only after the original execution finishes. The preview budget is set to a maximum of 45, or a minimum of 100, or 900 divided by the panel count. The full raw text is written within each job sandbox at /tmp/catalyst_primary_full.txt, but it is not assumed to be accessible next run. This fixes saved completion receipts, but not full document retention. The original scripts remain available in list_jobs ids 205 and 209 after cancellation. Full source URLs and branches remain in the original scripts and existing notes, but previews cannot justify trading. Local checks showed job 205 had seven print panels and job 209 had eleven print panels, and both reached the END without any reported errors. The commands used were bash pre91108/collector205.sh and bash pre91108/collector209.sh. The scripts executed the original Python body via exec with captured print, so there was no reduced source collection. The decision was to replace both jobs, but keep jobs 126, 132, and 172, and not add a sixth job.

A 44-minute plan starts at 08:01 UTC and finishes with a reserve in the last 5 minutes, with a hard stop at 08:46. Previously, near-term dates were known, but the expectation of a run-up was poor, so this will be set aside. The resolution will be based on the issuer's first news windows and the net executable stock payoff, rather than intraday headline noise. A question tree with 3 levels and 16 leaves will be used to make decisions. The tree includes factors such as unpredictable changes, including FDA status and TLX early decision, overlapping pipeline, and Zircaix news. It also considers overnight economics, including results from ORCL, ADBE, CPRT, and ZUMZ. Additionally, the tree looks at calendar completeness, including the next two sessions, and trade feasibility, including edge and structure. The plan starts with $49.14 in cash and no holdings, with the goal of creating an evidence-backed pre-open plan. The binding constraint is a positive executable expectancy. A strong countercase is new cash improvement and cheap positive skew, but new news cannot retroactively validate failed rules.

The fund manager of the Catalyst fund is researching IONS, MIRM, and GIS. The fund size is 0 and the current time is 2026-09-11T07:45:44.686421+00:00. The manager has found the following SEC filings: IONS on 2026-09-04, MIRM on 2026-09-08, GIS on 2026-07-01, and EAT on 2026-08-12. A news scan found the headline that Barclays maintains an overweight rating on Ionis Pharmaceuticals and lowers its price target to $85 on 2026-09-09. GIS had a conference on September 8, which has already passed. The manager will review the replay. GIS will release its earnings on September 23. The manager will review the earnings at 13UTC and will also review the Q&A session. The manager has set a baseline for GIS's FY27 EPS at 3.00-3.20, with organic growth at -1.5 to +0.5% and savings at $750m. The manager has also found that EAT will have an investor day on September 17 and an earnings call on October 28. The manager will review the earnings call at 16:30UTC. The manager has set rules for trading GIS and EAT, including verifying full pipelines and early news.

The Catalyst fund, which is for research only and has a size of 0, was updated on September 11, 2026, at 07:00:21 UTC. The fund is tracking several companies, including CNXC, which filed several documents, including a SCHEDULE 13G on August 14, 2026, and two SCHEDULE 13G/A forms on August 13 and August 6, 2026. A news scan found the headline 'Tamboran Delivers First Gas Sales From the Beetaloo Basin' on September 8, 2026, and two other press releases from Tamboran Resources Corporation on September 1 and July 20, 2026. The company also has a corporate calendar with upcoming events, including earnings reports on September 25, November 13, and May 13, 2026. Additionally, a news scan found information on NEOV, including a blog post on home battery safety on September 9, 2026. Other companies being tracked include TBN, SRRK, SCHL, and PRAX, with an exit value of 0.

A data scan found the following hashes for CNXC91106 data vintage: trades.json is d1eb7bb9443295af7ed57144607d41bf71a3d9a7572a94de1510e3069f174b7c, bars.json is 07db93734bf39c8b18584118c9cd77812f1c3354af5531a59fcd51ab926801ff, intraday.json is a9e2082aa9ce5206f5cef4d7b134c52a163aa958a04f7aa0b8820ea1ee43ad18, cli.json is fd3e962f2345b185f51e0f8a45fd52b054be1c60092492fd5565a38b564baadb, fundamental.json is bdce1e62e45200853548ad90cf4f2551e05f8cb958fa812dab42bf172948ed0b, events.json is 77551832b57c82c8f7238785a60a9b65b250d81e4de3ba236fcc05b1ea429375, intraday_results.json is 04181200f310b5ae265df03dd28380a1b5a5dca44ae7ae1e7542ae2cbfeea6ad, event_metrics.json is b6bb1ac50ec074a56381e40516a291acf5f631ab30a0bac29de53e3a6bb5be1e, results.json is a243377f3a1591ea906d78b9cd8d507e5ff97096ce7ae7208362e7eab1d428fa, and links.json is c6e326aab5d3071cce32b1e21d3d669e198f557464487977b1280dd06828dc8c.

The next run is at 08:00 UTC. The current cash balance is $49.14 with no open positions or orders. The existing monitor, number 250, is tracking NEOV and CNXC, with a review scheduled for September 22, but this is not a release time. The issuer date for CNXC can be found at https://ir.concentrix.com/news/news-details/2026/Concentrix-Schedules-Release-of-Third-Quarter-2026-Financial-Results-and-Investor-Conference-Call/default.aspx. Any new cash or financing improvements require validation of a new independent net event. Neither branch of the system is authorized to place an order. Any future positions in CNXC will default to exiting on September 28 during regular trading hours. No action is required before 08:00 UTC. Job 205 is already collecting CNXC SEC data, along with data from other companies. The system has receipts 132 and 172, but the outputs from jobs 205 and 209 are incomplete. All other catalyst monitors and jobs should be preserved. This analysis of CNXC is an extension of a broader screen, not a new whole-market analysis. The system has rejected certain daily and intraday data, and all expected bars are present. Event metrics are tracked from dates 24553 to 24575. The reproduction source is made up of chunks 2250 to 2260, and the calendar data came from browser originals, not empty links. Ledger 2679 measures the shortfall in stops, not the slippage in arrivals.

A Python script named intraday_study.py processed data and created a portfolio using the vbt.Portfolio.from_orders function with parameters such as size, price, fees of 0.0005, and slippage. The script then calculated the total return and number of trades, and appended results to a list called summ, which included values such as return, drawdown, sample size, win rate, mean percentage, and worst percentage. The results were saved to a file named intraday_results.json and printed to the console.

A news scan found the source code of an automated trading fund manager. The code is written in Python and uses libraries such as pandas, numpy, and vectorbt. It loads data from files named intraday.json and trades.json. The code then performs a series of calculations and simulations, using variables such as entry price, stop loss, and target price. It also uses a concept called slippage, which refers to the difference between the expected price of a trade and the actual price. The code runs multiple scenarios with different parameters, including slippage and a variable called be, which is set to either True or False. The results of these simulations are stored in lists and dictionaries, including out and summ. The code also calculates a net percentage change, which is the percentage change in the value of a trade after accounting for slippage and other factors.

A Python script named intraday_fetch.py is the source of this information. The script imports necessary libraries, including os, requests, json, concurrent.futures, and pandas. It loads events from a file named events.json and bars from a file named bars.json for the ticker CNXC. The script converts dates to the America/New_York timezone and formats them as strings. A function named get filters events by date, between 2025 and 2026-09-10. For each event, it finds the date five days prior and the day before, then makes a request to a website for stock bar data with parameters including symbols, timeframe, start and end times, feed, adjustment, and limit. The script uses a while loop to fetch all pages of data, then returns the event date, start and end dates, and the fetched bars. The script uses a ThreadPoolExecutor to run the get function on all events in parallel, then writes the results to a file named intray.json and prints a list of event dates and the number of bars for each event.

A script output shows a dictionary with keys including 'training_p', 'training_W_pct', and 'training_L_pct'. The output also includes a line of code that writes text to'results.json' and 'trades.json' files. Additionally, it prints a selection of data from the 'out' dictionary and two lists: 'PRIMARY' and 'SUBSETS'.

A news scan found the script output: the code calculates various metrics for a trading strategy. It first calculates the holdout root mean squared error (RMSE) and constant RMSE, which are measures of the difference between predicted and actual values, with the holdout RMSE being 100 times the square root of the mean of the squared differences between predicted and actual net values, and the constant RMSE being 100 times the square root of the mean of the squared differences between the mean of the actual net values and the actual net values. The code then calculates technical indicators, such as the simple moving average (SMA) and relative strength index (RSI), for a given stock, with the last date being the last date in the data, the close being the last closing price, the SMA20 being the average of the last 20 closing prices, the SMA50 being the average of the last 50 closing prices, the SMA200 being the average of the last 200 closing prices, the average daily range (ADR) being the average of the last 20 daily ranges, the average true range (ATR) being the average of the last 14 true ranges, the RSI being a measure of the magnitude of recent price changes, the momentum being the difference between the current price and the price 20 days ago, the realized volatility (RV) being a measure of the magnitude of recent price changes, and the dollar volume being the average of the last 20 daily trading volumes. The code also calculates the worst gap, which is the minimum of the differences between the opening and closing prices, and the SPY close and SMA200, which are the closing price and 200-day moving average of the SPY stock. The code then calculates the net reward to risk ratio, which is the ratio of the potential profit to the potential loss, with the potential profit being the difference between the current price plus three times the ADR and the current price times 1.0025 times 1.0005, and the potential loss being the difference between the current price and the current price minus the ADR times 0.9975 times 0.9995. The code then lists the trades that were sold, with the symbol being the stock symbol, the entry price being the price at which the stock was bought, the exit price being the price at which the stock was sold, the stop price being the price at which the stop loss was triggered, and the position size being the number of shares bought. The code then calculates the return, R, profit and loss (PnL), and stop slippage for each trade, with the return being the percentage difference between the exit and entry prices, R being the ratio of the profit to the potential loss, the PnL being the profit or loss from the trade, and the stop slippage being the percentage difference between the exit and stop prices. Finally, the code calculates the exit odds, which are the probabilities of exiting a trade due to a target being reached, a stop loss being triggered, or a time limit being reached, with the target rate being the mean of the reasons for exit being 'target', the stop rate being the mean of the reasons for exit being'stop', and the time rate being the mean of the reasons for exit being 'time'.

A source study found the following code: the function returns a value from an array. A list called ci is created to store information about the training and test data. The code calculates the mean and median of the net values in the training and test data, and the 95% confidence interval for the block means. The code also calculates the probability of a positive return, the mean of positive returns, and the mean of negative returns. A variable called hk is calculated based on these values. The code then runs simulations with different fractions and shock values, and calculates the median return, probability of loss, and other metrics for each simulation. The results are stored in a list called sims. The code also performs a regression analysis on the training data, and calculates the correlation between the features and the net return. The results are stored in a dictionary called reg. A news scan found the headline 'Descriptive training-only regression; no significance claims at N15.' The code continues with the line 'Xt=np.array([[x['mom20'],x['ma_distance'],x['rv20']'.